File: Screenshot 2024-07-19 013433.png (1.11 MB, 1000x800)

1.11 MB PNG

>mfw intel

i cant with these Canadian prices

OH MY

BUILDZOID REKT AND BTFO

https://www.youtube.com/watch?v=oicPuP-iDg0

>>101512149
>bug chasers
>redditzoid
pro tip: youtube is not an adequate way to gather information

File: 1455687281173.png (141 KB, 439x290)

141 KB PNG

>>101512309

>>101512149
Then where will I be spoonfed information? :(

tfw I just build a nice computer and be done with it. No need to fuss about oxidation, best of the best and the latest tech. You guys need to apply the same principle as if you were driving a beater car. It gets you from point a to b and it just werks. It broke? Oh well just buy another cheapass car.

If you invested 10k on Intel in Q3 2020 you'd now have 6.5k
If you invested 10k on AMD in Q3 2020 you'd now have 16.5k
If you invested 10k on NVDA in Q3 2020 you'd now have 77k

>>101512528
Is this why I keep seeing mfers in their teslas everywhere?

>>101511818
>recommended intel processors
>4070 ti super for 1440p
OP a faggot per usual

File: Screenshot 2024-07-21 190748.png (185 KB, 753x964)

185 KB PNG

>>101512149
>Thumbnail referencing P-Core VID not Vcore
It's like he knows the audience are dumb retards.
VID is worth monitoring only if the motherboard's CPU LLC setting is calibrated to match the ACDC load-line characteristics.
Very few motherboard manufacturers bother to set this up properly, they end up setting lower-than-needed ACDC load-line values like Buildzoid found out on his board samples.

That's still better than overvolting the CPU like Intel customer support suggests doing if your chip's on the verge of croaking.

https://skatterbencher.com/2021/11/25/skatterbencher-31-intel-core-i7-12700k-overclocked-to-5400-mhz/#Vf_Curve_Deep_Dive
https://www.overclock.net/threads/asus-maximus-z690-extreme-i9-12900k-guide-load-lines-vf-curves-adaptive-voltage-by-core-octvb.1794957/#Understanding%20LLC,%20AC_LL%20&%20DC_LL

The situation is well past this discussion anyway when -T SKUs with power-efficient V/F tables are degrading.

What's a good motherboard for a 7800x3d

>>101512862
any of them that meets your needs

>>101512862
https://www.youtube.com/watch?v=pUUVW7wgR3s
https://www.youtube.com/watch?v=57X2FygcqLE

These are my ATX suggestions if you're in the US, you can get a cheaper B650 board if you don't need the extra Gen 5 lanes.
It's hard to beat ASRock boards' price to available features.
https://pcpartpicker.com/product/Tz2WGX/asrock-b650e-pg-riptide-wifi-atx-am5-motherboard-b650e-pg-riptide-wifi
https://pcpartpicker.com/product/9BJp99/asus-tuf-gaming-b650-e-wifi-atx-am5-motherboard-tuf-gaming-b650-e-wifi
https://pcpartpicker.com/product/fCGbt6/asrock-x670e-pro-rs-atx-am5-motherboard-x670e-pro-rs

Gigabyte Eagle AX or Aorus Elite AX would be my B650 ATX board choices.

File: Screenshot_20240718_092348.png (31 KB, 1032x175)

31 KB PNG

>>101512149
buildzoid agrees with him?
>>101512788
LLC doesn't matter much in low-load boost
he does suggest using LLC 7 in his last video which is retarded

>>101513047
>buildzoid agrees with him?
CPU cores & ring aren't receiving the requested 1.5V voltage.
There's no point talking about Vcore at this later point of the community investigation.
Chips with lower idle & operating Vcore with saner factory fused V/F tables are degrading.
>he does suggest using LLC 7 in his last video which is retarded
What do you expect? He can't even stay up to date with the issue.

>>101512149
>intel's problem can be permanently solved by just locking the cores in bios to prevent single core boosting
>this will prevent any degradation
>if degraded already, also overvolt a large offset to stable (he did +0.5V to fix one that would bsod while idle)
Is he right? Serious question, because I had just bought an i9 14900K and I want to treat my hardware right. I assume undervolting will help for a new one that isn't degraded yet (and also save on the power bill).

Reminder to buy a 14900K and when it breaks buy another one.

>>101513133
All that babysitting won't matter if it turns out the chip has a manufacturing defect, which is inconclusive atm.
Go watch the GN video about it and determine whether your i9's part of the affected CPU batch.
There's not much undervolting room for the 14900K if you want it to run at stock boost frequencies and get the performance you paid for.

Just bought a 14900k. No problems yet, crash reports seem overblown

it crashed :(
buying another one

>>101513125
>CPU cores & ring aren't receiving the requested 1.5V voltage.
they're receiving something very close to 1.5 V, there's no significant Vdroop when only a single core is active
>Chips with lower idle & operating Vcore with saner factory fused V/F tables are degrading
what does the Vcore look like for a 13900T at 5.3 GHz? i assume it's shittier silicon, and the ring might start degrading at like 1.3-1.4 V
>What do you expect? He can't even stay up to date with the issue.
oh don't worry i understand he's only making these videos to appear competent to potential customers (and to get an opportunity to trash on AMD for no reason)
>>101513133
it's a shitty workaround but fundamentally yes
what you should do is look at the Vcore and frequency your CPU runs in a heavy load like prime95, and then lock your CPU manually to that voltage and frequency

>>101513212
>they're receiving something very close to 1.5 V, there's no significant Vdroop when only a single core is active
The factory-fused V/F table isn't set on a whim.
It's not enough to degrade or kill the CPUs given there's no manufacturing defect in the chips.
There's a sizeable portion of 13/14th gen i9 owners experiencing no issues at stock settings.

I'm not saying these chips are faultless, but people are focusing on the wrong angle here.
Even Intel's own CS reps are teaching their customers how to overvolt their degraded CPUs by a safe amount.
That's their first step to minimize RMA or refunds.

>>101513175
>There's not much undervolting room for the 14900K
In truth, I was considering a 14900T but those quite literally aren't sold by retailers. I got the K because it's the same hardware at different speeds for what ended up being the same price.
>-75% power consumption for -5% performance single-core and -22% performance multi-core
https://www.tomshardware.com/pc-components/cpus/intels-latest-lower-powered-cpus-give-ryzen-rivals-a-run-for-their-money

I was planning to imitate that as best as I could, to see if it actually pans out like a roundabout T. I also wanted the option to push the CPU the other way into overclocking and overvolting once CPU demands increased years from now (I didn't know about the degradation problem).

>>101512985
Appreciate it my man

>>101513276
>It's not enough to degrade or kill the CPUs
how are you so sure about that? intel might have hoped they would survive the warranty period and issues would have cropped up when the early adopters had already replaced their chips
anything to beat AMD, even if you're on a worse manufacturing process without chiplets
>Even Intel's own CS reps are teaching their customers how to overvolt their degraded CPUs by a safe amount
i would presume that, since intel hasn't put out any public statements, their CS reps haven't been trained to properly respond to this issue either

File: Thomas_Hannaford_0-171872(...).png (152 KB, 1514x776)

152 KB PNG

>>101513286
All you can do is set lower power (long & short duration), current, and temperature limits.
Stick with 245A current limit, the ICCMAX figure is recommended to help you get better Cinebench scores.
You shouldn't be feeding the processor over 300A of current when you're running more stressful like Prime95 on it.
Even some XOCers who receive test sample boards recommend setting a 125W long-duration power limit if you're going to torture test the CPU with stuff like P95 or y-cruncher.

>>101513311
>how are you so sure about that?
There are still plenty of people with perfectly functional 13th/14-generation i7s and i9s after it's been fed with over 1.5V throughout this time.
https://edc.intel.com/content/www/us/en/design/products/platforms/details/raptor-lake-s/13th-generation-core-processors-datasheet-volume-1-of-2/002/vcccore-dc-specifications/

>just disable ht goy
>just disable goy cores goy
>just undervolt goy
>just disable all the advertised features goy
>just dont run benchmarks goy
>just dont ___

>>101513435
>just dont ___
Talk about this at all. it is a non-issue, trust me. Forget about zen 5.

>>101513416
>There are still plenty of people with perfectly functional 13th/14-generation i7s and i9s after it's been fed with over 1.5V throughout this time
yeah, and they won the lottery and got better silicon that can handle higher voltages, or they haven't used their chips enough to cause any significant degradation
this is obviously why servers are failing first, they are active 24/7, but i don't imagine this use case was a concern for intel because they can tell server owners to buy xeon or get fucked

I have a rtx 1080 and comfortably play anythung I want

>>101513487
The worst instances of data center i9s have about a 50% failure rate according to L1, I wouldn't say the other half being reliable is "winning the lottery".
I'm not saying a 50% failure rate is okay, but you have very loose standards of winning a lottery with that survival figure.
And board manufacturers were setting lower than needed AC load lines, their stock settings were undvervolting the CPUs.
It still degraded for whatever reason and GN's source claims this could be a fabrication issue.

File: 13900KS_K_KF_comparison.png (186 KB, 3316x1906)

186 KB PNG

>>101513535
>The worst instances of data center i9s have about a 50% failure rate according to L1, I wouldn't say the other half being reliable is "winning the lottery".
there are huge variations in silicon quality on raptor lake and zen 5
these are complicated chips on tiny nodes
>And board manufacturers were setting lower than needed AC load lines, their stock settings were undvervolting the CPUs.
won't the boost algorithm compensate for this? that could be a cause of instability all on its own lmao

>>101513583
zen 4*
but zen 5 too, for sure

>>101513583
>there are huge variations in silicon quality on raptor lake and zen 5
>these are complicated chips on tiny nodes
The whole point of a VID requested from the individual CPU's unique, factory-fused V/F table is to account for those performance variations.
That is how ASUS motherboards derive a "SP Quality" score for the cores & IMC.

>Each processor is programmed with a maximum valid voltage identification value (VID) that is set at manufacturing and cannot be altered.
>Individual maximum VID values are calibrated during manufacturing such that two processors at the same frequency may have different settings within the VID range.
This is more likely to be a fabrication issue, and that is why there's a sizable chunk of functional i7s and i9s out there.

File: 66158.png (35 KB, 650x350)

35 KB PNG

>>101513643
>The whole point of a VID requested from the individual CPU's unique, factory-fused V/F table is to account for those performance variations
yeah, sure, i get that you trust intel to be competent and trustworthy but i just don't
if they expected these chips to see less usage than they actually did in the hands of users they could have lasted another 2-3 years with suicide voltages being applied to them
https://www.reddit.com/r/hardware/comments/1e09tml/comment/lcnrx0v

this isn't the first time a hardware company has released batshit insane products when they're behind the competition, none of them are immune to this

File: 1702870318928726.jpg (129 KB, 785x847)

129 KB JPG

>>101511818
what can you use a GPU for other than video games? i want an excuse to buy one.

>>101511818
what's the skinny on the 13900k?
been rending videos non stop for a while now with one, no crashes on the CPU side.

is that just a Gaymer issue

>>101513667
>yeah, sure, i get that you trust intel to be competent and trustworthy but i just don't
Intel's in a very costly manufacturing business and their objective is to sell products with low defect/return rates.
Their engineers aren't dumb enough to deliberately set unstable V/F tables in their costlier client products because they know better than anyone else in the world that it would accelerate processor degradation or outright kill them.
The situation is far from okay but it's more likely caused by manufacturing/fabrication defects than literal malice to screw over customers.

File: 1-11900KTurbo.png (27 KB, 760x476)

27 KB PNG

>>101513781
or, maybe, their engineers can't design a safe boost algorithm to save their lives
they keep on adding kludges to their decades old boost algorithm, i mean this shit still operates on the concept of "per-core turbo ratios"
AMD ported powerplay from radeon to ryzen and it works beautifully (sadly you can't modify the power tables, but they can be extracted if you know the addresses)

it seems too convenient to me that an issue which completely mirrors the symptoms of unsafe overvolting (faster degradation on i9s, faster degradation for 24/7 systems) would be caused by a "manufacturing defect" (which also conveniently didn't affect alder lake at all on the same node and with the same core architecture)

i'm giving my coworker my old pc but removing the hard drive, is there ANYTHING else i need to be worried about? like do I need to replace the ram as well?

>>101512577
There are no viable Intel processors right now.
There are no viable GPUs right now.

>>101514014
>There are no viable Intel processors right now.
12900k

>>101513884
It’s very easy to identify degradation caused by overvolted processors, yet no one was able to arrive to that conclusion.
That’s why it’s still an ongoing investigation.

>>101514014
>There are no viable GPUs right now
7900 GRE

File: file.png (101 KB, 1299x784)

101 KB PNG

>>101511818
2TB HDD is really slow even to open folders
these are the yellow warning signs in HWinfo

>>101514070
That shit's like 5 years old bro.

>>101513212
(and to get an opportunity to trash on AMD for no reason)
Buildzoid uses an 7800x3d and a 4070ti in his main rig wtf are on about

>>101514210
He's talking about FrameChasers, not Buildzoid.
The overvoltage theory isn't credible considering a fucking efficiency-focused 13700T with a maximum 4.9GHz P & 3.6GHz E core boost is degrading as well.
Even Alder Lake i7s don't need over 1.3V Vcore (a very generous figure) to sustain those boost frequencies, Raptor Lake chips have a better V/F curve.

>>101514210
framechasers says you should disable boost on AMD because "it's at 1.5 V!!! it's gonna degrade da CPU!!!" when that has been consistently disproven since zen 2
he wants to drag AMD down to the same level as intel even though there isn't a statistically significant amount of broken ryzen CPUs showing up in crash reports or at retailers

>>101514173
is it your boot drive?

File: 1721549862977731 rtx.jpg (453 KB, 1448x2048)

453 KB JPG

I'm gaming at 1080p and the GPU prices in my country are fucked, worth going for a 4070 Ti Super (980$) or it is a 4070 Super (780$) enough for at least 5 years of 1080p/60 fps/Ultra?

File: GS5c6SxXEAAMO2e.jpg (161 KB, 865x1968)

161 KB JPG

>>101513276
That's a false assumption
There are chip deaths even with 20-77 hours of 1.5v shoved into these chips
They seem to die from voltages that previous nodes could run

File: 71kIFMq+8dL._SL1200_.jpg (151 KB, 1200x1200)

151 KB JPG

Apparently pic related is a rebranded Seasonic Prime Ultra Titanium, and I ordered it on Prime Day sale for a real nice price. Did I do good bros? I will use it for my AM5 build.

>>101513583
Lol that graph is nonsensical
Comparing SP scores on processors with different frequencies is pointless according to Asus themselves
13900KS scores are inflated by ~5 points from it having a higher vf point 6 and 7 of 5.8 and 6ghz compared to the 5.6 and 5.8ghz of 13900K

>>101514419
>That's a false assumption
The majority of these 13/14th gen i9s deployed in data centers are still in operation.
Even the 50% failure rate brought up by Wendell claims half of that (about 25%) can be salvaged to a functional state, probably referring to either feeding more volts or lowering boost clocks.
12% of those CPU SKUs fail to pass GN's system integrator source.
The commonly accepted failure rate according to GN & Wendell's sources is between 10~25%.
The remaining 75~90% are running fine @ Intel's factory-fused V/F table.
You want overvoltage to be the main source of the issue, it's pretty blatant to see.

>>101514404
>Ultra
It's a meme setting most of the time.
If you're planning to own the card for over five years, get the best one you can afford.

>>101514467
https://twitter.com/jaykihn0/status/1815057866052751855

0.5% failure at dozens of hours is a pretty bad sign
Extrapolating this failure into the hundreds or thousands of hours that these CPUs are used in the field, the failure rate would easily surpass the "expected" 2-3% figure that older Intel CPUs have

>>101514507
That doesn't conclusively narrow the cause down to Vcore overvoltage.
No one here is saying the failure rate is fine.

>>101513738
It's a coin flip whether you're affected or not
Wendell literally claimed a 50% failure rate

File: 1699940444143102.png (22 KB, 92x92)

22 KB PNG

need a work station with at least 64 cores. budget £10k. UK

>>101514538
Get a reliable workstation builder in your region to sell you a good 7985WX Threadripper system.

>>101513703
Mining crypto, Rendering, AI imaging

>>101514556
i bought into threadripper 2990wx and then they swapped out the socket for no fucking reason so i couldn't upgrade, id rather go with EPYC or xeon this time.
i dont really care about clock rates that much, need cores in my case

Anyone else still using an ancient AIO?

I have an 120mm H100i from like 2014 that I have been using on AM3/AM4 since.

I remember people used to say they were good for like 1-2 years.

>>101514589
how do you get pcie 4 on tr4 socket though

>>101514589
>EPYC or xeon this time
>xeon
I'm pretty sure the next Xeon processors won't be socket-compatible with LGA 4677 boards anyway.
https://ark.intel.com/content/www/us/en/ark/products/237261/intel-xeon-platinum-8592-processor-320m-cache-1-9-ghz.html

A system using the newest 64-core processor will exceed your £10,000 total budget as well.

>>101514626
thanks, so threadripper is better in this case then? i guess i can overlook it

>>101514637
Yes, but a 7985WX will probably cost over £10,000 to build.
The 64-core chip itself costs $7,350 in the US.
The 5995WX is probably a more realistic option ($4,800 in the US, $6,500 launch MSRP).

>>101512149
>open video
>what ethnicity is that thing?
>bla bla gn bad bla bla intel good
>close tab
intel shill

>https://www.tomshardware.com/pc-components/cpus/intel-says-13th-and-14th-gen-mobile-cpus-are-crashing-but-not-due-to-the-same-bug-as-desktop-chips-chipmaker-blames-common-software-and-hardware-issues

it's joever isn't it?

>>101514467
>half of that (about 25%) can be salvaged to a functional state
Gimping the frequency and running the ram at 4200mt/s sure is great

>>101514673
No, out of all the YouTubers Linus cuck tips unironically had the best take. Intel will simply ignore the issue until it goes away, and that will probably happen when Arrow lake launches or when zen 5 launches.

>>101514190
Okay, would you like to tell me what home usecase needs better than 16 cores at up to 5.1 GHz?

>>101514580
rendering what

>>101514714
>16 cores
8 of them are running below 4ghz and are slow as balls

Thoughts on Acer - Nitro KG272U Pbmiipx 27" ? My cousin is asking me about it because someone recommended it to him and I cannot find anything about it besides the bestbuy listing.

>>101514737
Not an answer

>>101514619
No
But 'good' in this case means performing the same as day 1. I'm assuming you don't really use it on a 150w+ part either

>>101514368
no just a backup I dump vids & installers into

How small can a watercooled PC actually be?

>>101514911
You can set up a water-cooled PC inside an ITX case, but there's a learning curve involved (custom tubing & power cable management).
It's not recommended for relatively inexperienced PC builders.
https://www.youtube.com/watch?v=X0ukdo7Xx7U

>>101512528
Imagine not buying intc at this cheap valuation

Is NVIDIA's new app just an Electron app?

File: 1721536568389117 17208161(...).jpg (62 KB, 774x768)

62 KB JPG

4070 Super FE comes with PTM 7950?

>>101514404
the 4070 Super could do 4k/60fps easily without RT and probably could handle some UE5/PT stuff with moderate amount of DLSS
if you won't run RT see if the 7900GRE might be cheaper in your country, otherwise the 4070S + 1440p monitor would be a better choice than the Ti Super

>>101515266
>4070 Super FE comes with PTM 7950?
afaik only certain 4090s came with PTM7950 stock, anything under definitely ships with the shittiest and most poorly-applied thermal paste

>>101511818
suggestions on a prebuilt? i’m intimidated in building my own
i think microcenter will build it for me but idk

>>101515266
Get the 4070 Super FE for aesthetics only, AIBs make cooler & quieter models.
Check TechPowerUp & other reviews for more details.

>>101515357
Picking your parts and paying a builder's fee is probably the better option than buying a prebuilt.
System integrators will find corners to cut and still charge you a significant premium.

>>101515367
With the whole debacle about AIBs using cheap thermal paste that degrades really quick, I just want to buy something that doesn't require some tinkering in a short time span (and it voids warranty), this is why I'm asking if the FE cards from Nvidia come with PTM.
I've been reading some comments and there are people saying that RTX 4000 FE cards have pretty good cooling solutions compared to AIBs with no perceptible thermal degradation.

>>101511818
I changed the case of my pc but now the Post Status Checker shows a red light on Dram and CPU, the rgb of the mobo works, cpu fan works, gpu fans works, hdd works since I felt vibrations when I put my hand on it, the monitor shows no signal.
I had a mobo speaker lying around(I don't know if it works) but it makes no sound even if I take out all of the ram. I looked around for bent pins or something burned and everything seems alright.
Mobo is b550m steel legend.

What do?

>>101513899
The CPU works because the Israeli micro glyphs inside summon demons to do its bidding, so you'll need to exercise those before you give them to a friend

>>101514798
>Nitro KG272U Pbmiip
https://www.displayspecifications.com/en/model/ba803588
It's a budget VA, steer clear on account of black smearing and narrow viewing angles. For around the same price or less recommend your cousin the Dell G2724D ($180) or the ASRock PG27QFT2A (Was $160, but currently OoS) 1440p Fast-IPS options which are better all-around budget displays:
https://tftcentral.co.uk/reviews/asrock-phantom-gaming-pg27qft2a
https://www.rtings.com/monitor/reviews/dell/g2724d
https://pcpartpicker.com/product/vyqrxr/dell-g2724d-270-2560-x-1440-165-hz-monitor-g2724d

>>101515454
Much appreciated! I saw that Dell in the OP list.
Which of the two do you recommend? I'll send him both, but some differences would be nice to quote.

File: Screenshot 2024-07-22 005903.png (775 KB, 1386x1013)

775 KB PNG

>>101515411
>I've been reading some comments and there are people saying that RTX 4000 FE cards have pretty good cooling solutions
I own a 4080 Super FE and its cooling capacity is the least I'm impressed with, note that the cooler is designed to handle the 4090's 450~600W heat load.
I can't even get the card to hold a 2900MHz boost frequency at stock fan curves, the one set by NVIDIA to optimize operating noise.
My 4080 Super GPU is one of the best in an FE card according to 3DMark results, but the cooler's throttling performance.

Good day. I've got this old rig with an i5 4460 on it, thing is the CPU would heat up quickly doing some simple tasks, I cleaned the paste and reapplied it(thing still has its original cooler) and worked for the most part, now it only really heats up(like up to 90C) when playing video, like playing a 1080p video on YouTube(specially if you use the arrow keys to navigate), would installing a cheap video adapter like the GT 1030 appease that or would it be a waste?

>>101515477
>Which of the two do you recommend? I'll send him both, but some differences would be nice to quote.
One thing is the Dell does come with an sRGB gamut clamp while the AsRock doesn't which means OoB colors in SDR (normal webpages, desktop, non-HDR video) it should look more accurate but ICC profiles for both displays can be downloaded from reviews to somewhat rectify stock color performance.

We don't have a comparison review that pits them head-to-head yet so any performance differences should be taken with a grain of salt. Both measured ~1400:0 static contrast and ~430 nits max brightness in their reviews which is above-average for IPS (normally in 300 nits and 1000:0 contrast range). The Asrock board has a slightly wider measured gamut (93 vs 88 % of DCI-P3) and it's refresh rate is higher (180 Hz vs. 165) while both have a roughly equal low-6 ms G2G response at their max refresh rate and Overdrive Settings.

>>101515155
INTC has literally nothing going for them lmao

>>101515568
Wonderful. He won't go beyond plug-n-play, but it's good to know that stuff. Thanks again.

>>101515576
NP

>>101515432
>the monitor shows no signal.
You did plug the monitor into one of the display outs of the GPU and not on the motherboard right?

is there a reason why people ridicule you for using userbenchmark?

>>101515432
I didn't touch the cpu/ram before I started troubleshooting the problem, the only thing I did in relations to those 2 components is grabbing the mobo by holding the stock ryzen cpu fan.
>>101515589
Of course, but even if I didn't the pc wouldn't boot with red lights being active on the Post Status Checker, no?

>>101515602
It's run by a blatantly impartial webmaster.
It's like getting sex advice from gay men, they'll tell you pussy's bad and you need to suck more dicks.

>>101515622
I meant to say partial but I am a dumbass.

File: Damned Lies.png (2.09 MB, 1919x3330)

2.09 MB PNG

Fucking hate these shady ass "sales". Does anyone know of an easy way to find fake sales? The wayback machine is useful, but it's not completely reliable to get a complete record of an item's price over time.

>>101515663
$300 my fuckin ass

>>101515663
camelcamelcamel

>>101514230
My bad man i should have realized, i still don't trust that Framechasers guy though i just watched the video and it literally is just him rambling away

>>101515663
atleast it got dram not hmb cukc

>>101514155
>sips on 100W while watching goytube

>Intel CPUs Are Dying, It's Over (featuring Future Anthony/Emily Of Level1 Troons)
>I9s Are Rusting And Intel Needs To Say Something
>Intel Bad
>9950X Overclocking (it died after 1h of oc'ing)
>Why The New Poozen Architecture Is Good
>Buy AMD CPUs Now

how long can i ride out am4 with a 5800x3d boys

>>101511818
>how would/g you consider this pc for a poorfag? Taken from an Amazon shop


Case: Sharkoon TK5M 2xUSB3.0 VETRO TEMPERATO,4X120,ARGB FAN - Motherboard : B450M (GigaLan, Audio, USB3.0)
- CPU : AMD Ryzen 4500 6core 12Threads up to 4,10Ghz
- M.2 512GB nvme PCIe
- Ram : 16GB DDR4 3200mhz
- Nvidia RTX 3060 12GB
- psu 600W 80 Plus
- Windows 11 Professional

>the idea is to use it mostly for 1080p gaming/simracing/ video and photo editing.
the 3060 would also allow in my plans for decent sized stable diffusion generations.

>>101515754
10 years at least

>>101515432
Anyone? Should I try to get them (mobo, cpu, ram) repaired/replaced under the active warranty?

Guess us europoors at least have the 2 year warranty going for us?

>>101515884
Lousy so commoners, maybe post a picture. Might be you put it together wrong.

>>101515762
the 4500 is bad value, try to aim for 5500/3600

File: 1718970206477450.jpg (65 KB, 910x1024)

65 KB JPG

>3600x
>2x8gb 3200hz
>6600 xt
>325 euros total
I just wanna play some modded assetto corsa

>>101515978
the pc is good but you should play beam ng instead

>>101515978
its a 10 year old game. itll run on a toaster

>>101515974
I'm currently on a i7-6600, would it be any upgrade?

File: 1609865807665.jpg (79 KB, 500x579)

79 KB JPG

>>101515097
>turns 4090 literally into 4080
L M A O
SFFtards are just on another level of retardation.
This dude with mclaren in his room can afford 4090 every month I guess.

File: 00519-2167690355.png (1.62 MB, 1152x1728)

1.62 MB PNG

I desperately need an rtx 5090 so I can generate more 0 effort ai sloppa

Please nogvideo release it in September....

With 32gb of vram yes yes

File: 1685918223200.png (134 KB, 480x360)

134 KB PNG

>>101516016
Take a sip, lifeless loser.

>>101516001
Yes. but look into 5500 and 3600 it's only $10 more

>>101516030
Sip of what
I'm dying here this 3060 is so slow and 4000 series is a fossil at this point

what exactly is the deal with the melting 4090s? Is it only when using the adapter or also when using 12vhpwr directly?

My understanding is that they melt because they are incorrectly installed (the cable not inserted thoroughly) so I triple checked my connection and it seems alright. Will it melt regardless? I also undervolt and only connected three out of four cables.

Does getting a new PSU with 12vhpwr reduce the chance of melting?

>>101515898
>maybe post a picture. Might be you put it together wrong.
What could I possibly fucked up if all of those happen?
>the rgb of the mobo works, cpu fan works, gpu fans works, hdd works since I felt vibrations when I put my hand on it, the monitor shows no signal.

>>101516226
Getting a PSU with a native 12vhpwr will only make it so that you now have a risk of it melting at the PSU end
The best compromise would be to run an atx2.0 PSU with a direct 12vhpwr into the GPU that splits into 2x or 3x 8-pin pcie at the PSU side

>2024
>Faggots on 4chan still pretending to hate the 4000D because it’s popular
You contrarians are insufferable

>>101516226
4090 have 2 ways of death:
a)melting connectors
b)melting drmos
ahaha, nshitia consumers are utter retards at this point.

>>101516238
take the board out of the case and run it that way

If i want to apply ptm 7950 for my 79xtx, should i also buy and replace the thermal pads for the memory chips, or will the stuff still work fine after remounting the cooler? The merc310 xfx model has a rather high hotspot and higher fan rpm makes it sound like a ps4.
Addendum; is water cooling the gpu worth it? The waterblock alone is like 170€ without the pump, rad, reservoir, etc.

File: 002ueA76ly1guc0k8d9gdj631(...).jpg (189 KB, 1280x1920)

189 KB JPG

Anons, is there a way I can make either my GPU or monitor display a blue-only image? I'm trying to calibrate the color with the AVS 709 patterns but I need a blue-only mode.
(Better calibration methods are appreciated as well)

>>101516638
Yes to the pads

Philips 27M1N5500ZA/00 currently 210 euro shekels here. Are there any better choices around that price range?

>>101516671
i would have told you if you didnt post such an ugly creature
look at its face it gives me the creeps

>>101516638
A bad repadding can cause catastrophic damage. Too thin and you lose contact, overheating, too thick and you warp the PCB. Find out exactly how thick the various pads are and order 1:1 replacements before disassembly.
Custom watercooling is not worth it in general for daily use but it's a fun project if you're okay with the maintenance.

>>101516749
>Philips 27M1N5500ZA/00
https://www.displayspecifications.com/en/model/7eda28c9
That looks like an LG Fast-IPS rebadge.

Where are you located in the EU? In DE for example, you can get a slightly different LG panel (165 vs 170Hz) for €20 less (performance differences otherwise on these Nano-IPS panels will be negligible tweaking):
https://de.pcpartpicker.com/product/2FmNnQ/lg-ultragear-27gr75q-b-270-2560-x-1440-165-hz-monitor-27gr75q-b

>>101516749
Fairly standard price for a budget NanoIPS. If that's what you want then I wouldn't say no.

File: d5592dd26218fb939c3c0dc81(...).jpg (269 KB, 2000x1985)

269 KB JPG

>>101516808
Now tell me

>>101516867
eastern europe, gr75q is 199€
Gigabyte g27q around 210
Gs27 210€

why do people go for intel for amd top cpu like 1900k and 7900 when most of what these people do is web browsing discord and gaming
>work
lol

>>101517008
if the 16 core cpu costs like <500€ and you do productivity stuff on the side you might as well pay the extra 30% so your programs finish in half the time

ofc budget oriented gamers arent buy >8 core cpus, and gamers arent buying intel flat out since the 3600 came out

>>101516808
Yeah I don't know why chinese cosplayers overdo the photoshop, it's really offputting
It wouldn't be nearly as bad if they would just stop samefacing

>>101517008
"futureproofing"
And mostly because they can easily afford it, what's another 200 dollars if you're spending 2k+ as it is

>>101516226
It's literally retards not fully seating the power connector or using that shit 3rd party right angle adaptor

Should they have been made more idiot proof? Yes
Is it a design flaw beyond that? No

>>101516956
The GS is a cheap VA which pulls it out of contention on account of dark smearing.
Compare the gr75q & g27q measurements here:
https://www.rtings.com/monitor/tools/compare/gigabyte-g27q-vs-lg-27gr75q-b/12904/44761?usage=3623&threshold=0.10
The long and short of it is the LG seems a bit faster in refresh rate and response times, but the Gigabyte model gets significantly brighter, has better color volume and gamut, and better uniformity (less lightbleeding).

If it were me I'd pick the G27Q.

>>101516671
Monitor service menu
If it can't then it's shit and not worth calibrating

File: 1719835440111923.png (81 KB, 198x245)

81 KB PNG

>>101517045
>>101517027
thanks
get a (you) counter increase

Is PNY a good brand for GPUs or should I go for a more known brand?
Want to buy the 4080S

Best leaker to follow for GPU release dates?

Do companies really wait for events like CES to announce their next products?
I want next generation of GPUs NOW

>>101517231
always do a google search with warranty after the brand
reading a reddit thread is seems pny bitches about paste change and wont refund meanwhile msi seems fine with that and chill

Damn, my shits not stable with XMP on. So far it's only stable at default settings which is 4800CL40. Currently testing with the Asus OC profile which the board defaults to which mentions dynamic memory timing. Only an hour in so far but every other setting failed after a few mins.

>>101513703
I believe folding@home works on gpus.

>>101517313
which cpu and board sir

>>101517325
it's tuf gaming B760m-plus wifi II and a i5-14500
ram is corsair 64 Gb 6400CL32

lots of ram in the QVL list running faster than that on 13th gen Non-K processors apparently...

>>101517313
Full specs?
CPU, motherboard and ram model code

>>101517363
give it some more system agent voltage

>>101517363
Is your CPUs stepping B0 or C0 as shown in hwinfo64 or cpu-z?

>>101517363
try returning the cpu and buying amd instead

>>101517378
it's C0

>>101517391
That's weird, it should just work then
I guess your CPU's memory controller is just bad

File: IMG20240122124924.jpg (3.82 MB, 4608x2592)

3.82 MB JPG

>>101516638
If you are careful and treat it like your firstborn during disassembly the pads can be re-used.

>>101517391
Anyway you can either try to return your CPU and get another 14500 that may work with it, return your memory kit and get 6000cl30 or just live with the jedec timings

>>101517367
>ram model code
CMK64GX5M2B6400C32

>>101517370
I'll try some more manual settings i guess, just disappointing it's probably not going to reach it's max

>>101517401
Yeah could just be bad silicon lottery luck

>>101517420
Yeah I mean these chips don't benefit that much from overclocking the ram anyway. A few percent at most.

I'll do some more testing, see what I can get out of it and have a think about it. The slower ram isn't much cheaper either.

Can I get some monitor recommendation for around 300 euros? A lot of online resources and discussions are US centric with model recommendations that either aren't available here or are quite a bit more expensive than in the US.
One that's on my radar is 27GP850-B.
Dell G2724D also seems very good, but it's slightly more expensive and has limited availability with long delivery times (Amazon says 3 weeks). With the LG i can go to the store right now and get pick it up in 10 minutes.

is diy egpu enclosure an option

File: 1710532662557606.webm (1 MB, 576x1024)

1 MB WEBM

>>101515995
no weeb tokyo larp in beamng
>>101515998
apparently mods+traffic is hard to run good

Also is 500w supply enough for it? got 3 small SSDs currently, gotta check what my wattage is on the psu
>>101515978

>>101517716
You can do it. Should you is a different question entirely.

>>101517688
What country (or countries if you can cheaply import)?
>One that's on my radar is 27GP850-B.
>Dell G2724D also seems very good
https://www.rtings.com/monitor/tools/compare/lg-27gp850-b-27gp83b-b-vs-dell-g2724d/24452/41041?usage=4112&threshold=0.10
For unique features, the LG has backlight strobing and the Dell has PS5 VRR support.
The LG will be faster in motion and have better color gamut but worse black uniformity and below-average contrast for an IPS which was a key downside to the earlier-gen NanoIPS models. The Dell won't be amazing at any one thing but alright at everything without any major weaknesses.

>>101517688
https://www.rtings.com/monitor/reviews/aoc/q27g3xmn

or odyssey g6 perhaps, it has screenlines tho
avoid IPS

why cant nvidia get good software like amd adrenalin so i dont have to rely on this msi afterburner and rivatuner bloat that randomly stops working so i have to leave it open 24/7

why cant geforce experience not be a bloated shitheap when all the core functionality is on the nvcp as its been for the past 20+ years

but nvcp has no native clock, voltage, or fan control and geforce experience is bloated trash AND forces you to log in which means i have to use 7z to extract driver from package and install through device manager and obs replay buffer scripts instead of shadow play

people talk nvidia up about its drivers being the best but they dont seem able to tie it together for shit

>>101517777
i have an old ivy bridge thinkpad and an old 1060 i could get some mileage out of with a good screen but the enclosures cost more than what both are worth combined so theres no point unless diy is significantly cheaper

>>101517826
have you checked out the nvidia app beta

>>101517843
complete waste of time and money
the money you spend on the egpu doodad is better spent on a cpu/board/ram combo
harvest the laptop for a drive
boom you now have a gaming pc

File: 1721084600451355.webm (258 KB, 436x304)

258 KB WEBM

>>101513212
>they're receiving something very close to 1.5 V
Fucking why.

>>101517859
looks good in cpu-z benchmark

>>101517859
The average 14900k needs 1.498v for 6ghz boost and 14900ks' need 1.55v for 6.2ghz boost
That's before factoring in AC Loadline that'll push those values up by 50-100mv

>>101517850
i have not but if it requires a login and bloated telemetry and geforce experience package to use basic functionality ill stick with my scripts and suffer the afterburner and rivatuner overhead instead which will be no doubt less than whatever horrifying kafkaesque appstore horror show theyve cooked up to hoist on the windows IPR mind slave appstore™ biomechanical botnet

>>101517899
doesnt require login unless you want to redeem rewards or something

anons bros are these tweaks worth it?https://www.youtube.com/watch?v=EEEKtsW2ZzE

>>101517856
i already have a gaming pc with a 4080, a server, and 3 other more antiquated gpus to troubleshoot slots with

>>101517894
I have 13600k and at LLC 3 I get 1.19V in C23 for 24k points. Maximum reading is at 1.25V which is not even VTOUT so the actual value is slightly lower. Is there such a drastic difference between these chips that you need to blast it with 1.5V? I am not even surprised the i9 line of 13th and 14th gen is shitting itself if true.

>>101517910
if you like less stability and 5 more frames then sure

>>101517910
Without watching the video I can tell you there is a lot of bad advice on youtube. If you want to follow the video, you should check out every single setting and what does it do. I'll give you an example for Intel. If a video is telling you to disable c-states to get higher performance, if you disable c-states you also lose maximum turbo boost for performance cores because the CPU is not designed to run P cores at turbo together with E cores constantly.

File: 71S-ID2SkiL._AC_SL1500_.jpg (89 KB, 779x1500)

89 KB JPG

>>101511818
Hello, I want to buy a wireless network card and narrowed it down to these two:
>https://www.amazon.de/dp/B0BZYRPG4J
>https://www.amazon.de/dp/B0C9PPRPM3
The thing is they look exactly the same, which being chinesium is not surprising. but ARE they the same though? As in, is there a chance one of these companies has a better (or less worse) version of this same product? e.g. by virtue of the other company using a cheaper material

>>101513155
>what is RMA

>>101517909
this appears to just be a bloated nvcp wrapper with no extended functionality as far as i can see

i dont care for omniverse, afaik you need a login for that other stuff too.

>>101517937
It's what intel programs these CPUs for and what shows up in Asus' bios
The vf curve gets really steep past 5.4ghz so you won't feel the effects on an i5 unless you go for max clocks.

>>101517994
For example, the first one says (translated from German):
>Contents: 2x 6 dBi Antenna, 1x Low Profile Bracket (8 cm | 3,15 Zoll), 1 x USB Cable (30 cm | 11,8 Zoll), 1x Screwdriver, 2x Screws
While the second says:
>Contents: 2x 5 dBi Antenna, 1x Low Profile Bracket, 1x Bluetooth Cable, 1x Screwdriver, 2x Screws, 1x User Manual, 1x PCIe AX210 WiFi-Card

Obviously the first one also comes with the card, but one says 6 dBi while the second 5 dBi antenna. Is that trustworthy? As in, there is a difference, or is it just a typo or bad copy pasting?

it's almost as if Intel doesn't care about quality or users and only wants to pad their Cinebench scores.

>>101512482
basically my philosophy, drive a basic Honda civic and upgrade my GPU every 5-7 years, save and invest the rest

>>101517910
no

This my list; https://au.pcpartpicker.com/list/j8NfQP

Already bought the CPU, just need to buy the rest. Was wondering how much is the difference between 4080 and the 4080Super. The benchmarks I saw showed minor improvements.

Any advise?

>>101512482
>>101518227
But for what you know it might not be best for saving money. Sometimes its better to spend more one time than spend less multiple times as it will save you more money in the longer run. Go read boots theory on wiki and you will understand what I mean.

>>101518264
they're effectively the same card, get whichever is cheaper

>>101518264
MSI is sussy

>>101515445
but he also needs the CPU to work

How necessary are those rubber pads on case fans? Should I avoid ones without them or does it not matter?

>>101518264
decent build
don't bother buying extra thermal paste like i did, thermalright includes a tube alongside the cooler

>>101518286
it helps reduce noise.
if your fans don't have the rubber pads you can cut up a bicycle inner tube or shove toothpicks in there

>>101518286
Not massively - they just exist to reduce vibration (and by extension, noise) between the fan frame and the case.

>>101518276
yeah you should totally get asus aka fry your x3d cpu and scam your rma
or maybe gigabyte with their cracking pcbs

>>101518301
i would pick Asrock, seems to be the best this gen

>>101514404
6800XT will set you free five years lil nigga, will easily handle ultra and high at 1080p in 95% of games, ignore AAA slop

RTX 4070 Super Gigabyte Eagle OC or Gaming OC, which is better?

>>101518322
Without looking anything up, I'm going to say Eagle is better because the name is fancier

>>101518264
Super is marginally stronger and on average cheaper.
Get an nvme instead in this budget.
>>101518276
I have that mobo, what is supposed to be "sussy" about it?

>>101515754
5800x3D and 3080 10GB here, gonna ride this bitch to 2028 minimum

>>101512149
this nigga genuinely adds nothing to the discussion

>>101518322
Gaming is marginally better build quality, but only buy it if it's like 3% more expensive max than the Eagle. They're both great 4070 Super cards.

>>101517799
That AOC looks tempting. But it's 100€ more expensive than the LG and the viewing angles of VA are really off-putting.

>>101518330
I like its design desu.
>>101518347
I see. They're 615$ and 625$ respectively. What about Aero, I can buy it for 599$?

>>101518360
>What about Aero
Aero is Gaming OC with a paintjob. Grab it if you don't care about the color; I've even seen it sold quite a bit below MSRP a while back, all because people care too much about muh aesthetics.

what key do i need to press to open bios in msi

>>101518360
Aero is normally pricier than the gaming oc because of the white tax so you're in luck

>>101518406
Nearly every bios opens by F2 or Del.

>>101518406
just spam both del and f2
you have 2 fingers right anon?

>>101518406
f11

>>101517040
otherwise, you'd see their hideous bug faces
Gooks go even further and photoshop their shit in real life

>>101518427
>white tax
Chuds were right.

>get new gpu
>comes with starfield for free
This game fucking sucks, I wish manufacturers let you give away the keys they give you. glad amd got their shit together at least.

>>101518332
>Get an nvme instead in this budget.
What do you recommend? Western Digital Black SN770 2TB?

Not a big fan of nvme. Heard something about shorter life spans for nvme some time back.

File: 1718203215295736m.jpg (118 KB, 1024x833)

118 KB JPG

Is the 4070 ti super worth it? I see people saying to just get the 4080 super, but I'm already pushing my budget with an $800 card

>>101518476
Consider sn850x instead. I am on kc3000 personally.
>Not a big fan of nvme
Why? Lightning fast compared to sata(although won't matter too much for games).
>Heard something about shorter life spans for nvme some time back.
My current one is working pretty fine.
My previous nvme still works after 5 years.

PC scrub, looking to buy something like this: https://pcpartpicker.com/list/t3sc89

Not sure on a monitor. I had an Asus XG27AQV monitor picked out, but kind of second guessed myself. I originally wasn't going to get a 4070 TI S, but was convinced to spend the extra cash.

https://wccftech.com/former-samsung-engineer-jailed-for-6-years-leaked-oled-information-to-china/
OLEDs for $300 soon. Hooray industrial espionage!

>>101518489
I'd get 7900GRE, but I'm the case of "works on my machine, always"

>>101518489
Worth it if you:
Are interested in AI.(can fit larger models than 4070 super)
Are playing recent AAAs on high resolutions(extra vram for textures)
Love ray tracing(extra vram)
Otherwise just 4070 super if you are on budget.

>>101517688
try searching a bit deeper, I doubt there isn't a sub 300 G2724D for you anywhere, unless it's perpetually sold out
price trackers show the monitor to be even sub 250 at some times for me, but I just bought an extra one yesterday from seemingly the main dell representetive in my cuntry (dellshop.lt)
mostly around 300 from the popular sellers though

>>101518265
Yeah you definitely want to BIFL on the stuff that interfaces with the real world. With a PC you know for sure that it'll be obsolete in a few years so going cheap on the "computing" parts is the play. Of course "cheap" is relative - if you make 200k then building TotL is fine.

>>101518589
So it's really only worth getting if I need the 4gb extra vram compared to the 4070 super?

>>101518604
My rule is spend 10% of my savings on quality of life(PC usually), if I can cover it with interest in a year. Lose on inflation, sure, but I don't buy PC parts every gen, I skip one or two.

>>101518648
Raster difference isn't enough to justify 200 extra dollars so yes.

Kraut anons, does anyone have any idea how long it takes for G.Skill to process their RMAs? My ram kit which I bought from Computeruniverse became defective, so I sent it back about three weeks ago, and the faggots at Computeruniverse told me that my ram kit arrived to whoever they sent it to last week.

If it's going to take a few more weeks, I'm thinking about buying a cheap 16GB DDR5 ram kit from Amazon and then returning it after 2 weeks.

>>101518476
It's slightly complicated. On the NAND side an NVMe has the advantage because the interface is designed for SSDs. Less write amplification and HMB reduces FTL writes. However, for consumer use, NAND durability is seldom a problem - the controller is far more likely to die before you exhaust 1000+ TBW. Here SATA drives have an advantage in theory because they're so slow that they don't really get hot. Again though consumer applications don't involve thrashing the drive with 100GB writes so it's probably a wash. In any case, the solution is to use a heatsink; not pay more for less with SATA.

>>101518563
>My current one is working pretty fine. My previous nvme still works after 5 years.
I'll research some more.

Is Crucial or Samsung good brands for M2? or stick with WD?

>>101518733
research controllers not brands
usually one year 80% of NVMEs have exact same controller and it's wise to get a cheaper version

>>101515703
>turns off hardware acceleration in your path

>>101518733
Samsung is good but overpriced.
Just stick with the two I recommended to you and you will be fine.

>>101518757
hilarious part nvidia users have to turn it off because youtube flickers black mosaic at them

also my 7900xt eats only 40w watching 4k youtube

>>101518681
>the controller is far more likely to die before you exhaust 1000+ TBW
You've reminded me what the issue is now. Upon reaching total write allowable, the start having corrupted data issue.

>Again though consumer applications don't involve thrashing the drive with 100GB writes so it's probably a wash.
Not to sound stupid, but, as long as the PC doesn't get too hot I should be fine? Or is the assumption that the chance of daily 100GB+ of data is unlikely for the average person?

>>101518857
>the start having corrupted data issue.
my 840 EVO 250gb is 12.5 years old, I used like a PB at this point, seem to work alright

>>101518857
>Upon reaching total write allowable, the start having corrupted data issue
That's all SSDs not just NVMe or SATA. You generally don't want to fill them beyond 80-90% and obviously once the TBW is exhausted you'll need a pro to retrieve the data. Don't do that though - replace the thing when it hits 70%. They're not expensive.

>As long as the PC doesn't get too hot I should be fine?
Well if you're going for maximum longevity you'll want to keep the SSD controller itself cool. Most of the time that just means using the included motherboard heatsink but, if you are actually writing 10s of GB per day, an aftermarket one would be a good idea. Rare for a consumer to do that.

>>101518750
What to look for in controllers?
nvme are brand new for me. Last time I built my PC was about 10 years back, and wasn't able to keep up with progress due to work.

no other thread to ask, which laptop should I get for a game like maplestory? I have a pc but I need something mobile, battery power isnt important because it will be plugged usually

>>101518954
https://en.wikipedia.org/wiki/List_of_flash_memory_controller_manufacturers
Phison is usually go to. Try to avoid chinese controllers, it's a roulette, chinese brands are alright if it isn't fake, a lot of NETAC fakes around from what I heard.

>>101518947
>You generally don't want to fill them beyond 80-90% and obviously once the TBW is exhausted you'll need a pro to retrieve the data.
Wouldn't it be better if an HDD was used as primary/long term storage?

Is CPU overclocking still relevant? i see people mention it much less compared to like 6 years ago

>>101518996
any modern APU can handle it

>>101519011
No, these days amd and intel boost the cpus clocks so much out of the box they degrade and die early or cause massive heat spikes that require your cpu cooler go max fan speed sporadically which is super annoying

>>101519011
these days your cpu is factory overclocked to the max from when you open up the packaging, and it's your responsibility to underclock and undervolt it to usable levels

>>101519011
no, OCing GPU isn't relevant either
chips run at max efficiency at this point, doing manual OC only makes them hotter and pwoer hungry for miserable performance icnrease

>>101518998
Gotcha. Cheers mate.
I've now got some research to finish up tomorrow or day after.

>>101519029
in the case of intel you can even buy pre-underclocked cpu mobo ram combos from bugchasers dot com

>>101519011
Only undervolting because everything gets hot as fuck and dies faster.

>>101519038
haha, someone took my idea seriously, no pre undervolted GPUs yet? Can't wait before Asus comes out with eco friendly 7090 or 9900XT
"-15C' and draw -50% power for only 10% performance!"

>>101519071
with gpus its just a couple clicks with msi afterburner

>>101518779
that's not a youtube thing it's a chromium thing

File: 1721652784030.jpg (3.15 MB, 4000x3000)

3.15 MB JPG

Why people in Russia are so obsessed with water cooling? I got mine 4080+7800x3d PC purely on air (assassin iv) and people are telling me that I'm poor.

>>101519095
its not a chromoim thing its win10 thing
win 11 fixed it already

>>101519008
Kinda depends what you mean by that. If budget isn't a big issue then these would be your best options:

>Primary OS drive
Optane/3D Xpoint
Realistically, NVMe (TLC w/ DRAM or HMB)
>Secondary game/data drive ≤4TB
NVMe (can be QLC) or SATA SSD if slots are limited.
>Tertiary/warm storage >4TB
HDD. Ideally in a NAS.
>Cold storage
Tape drive or M-disc.

>>101519099
хpeн их paзбepeт, нa мeня кaк нa дypaкa cмoтpeли в 2017 кoгдa я Hoктюa Д15 бpaл зa 7к, a щac типa фигня, y нac в poccиях caмoбытный быт

>>101517994
>>101518067
In case someone browses the thread and wonders: Went with the first one just in case the antenna is indeed (even minimally) superior.

>>101519099
Normalfags are just retarded

>>101519095
same thing.

File: 1644971670816.jpg (257 KB, 648x1003)

257 KB JPG

Day 473 of waiting for next-gen GPUs.

>>101519099
Water cooling is something people do when they want to seek attention. People who are spiritually poor love to seek attention. Eastern Europeans (like Buildzoid) lack soul.

>>101519099
>>101519131
I'm surprised anyone in russia can even afford such things. Anytime I play online with russians vast majority of them are: laggy af whoping 20 fps and 200+ ping, can't run on lowest settings, probabaly run DDR3 or even DDR2 generation junk that unironically have been passed to them from their father or even grandfather at this point. And of course creme de la creme the microphone that predates xbox 360 era and was used during unreal tournament in 1999. I have nothing against them but having to use 20+ year old computer tech today must be depressing af.

>>101519011
It's not
To add insult to injury, the fastest gaming CPU is a locked processor (7800x3x)

>>101519219
only children play online games here, adults got tired of that shit in mid 00s, WoW is kind still popular among 30+ I think
I got 7700x/7900XT Red Devil/32Gb RAM+ Fractale Torrent+ D15+ and NVMes and I got them real close to US prices except GPU that one was like $950 in local tugriks. Plan on OLED monitor soon, and I'm not rich at all.

>>101519317
oh remembered a nuance, you wouldn't know your are playing with a russian most of the time btw, we hide well, again only kids speak russian online like everyone else must understand them

>>101519215
>Eastern Europeans (like Buildzoid)
Where do you think Buildzoid lives?

>>101518604
I got my 2017 gtx 1070 and it still runs modern games. If by obsolete you mean that it can't run highest settings then yeah everything will go bad quick but I bet I could run my gtx 1070 a few good years if I don't mind playing low settings and just 50-60fps on brand new releases. I probabaly could squeeze more with some optmization of my system, closing all uneccesseary apps maybe even adding extra 2 ram sticks for cheap. If I were to buy gtx 1050 instead which back then was like 100euro less then I probabaly couldn't do that and needed an upgrade now.

>>101519402
>and it still runs modern games
at 1080p, at 1440 it's dead.

>>101519219
same as thirdie SEA, cheap cost of living

gaming are cheap hobby.

>>101519317
>>101519219
Yeah for some reason EU poorfags are obsessed with WoW. They flood private servers to the brim. Their computer can't run retail because the modern client is too bloated hence the need for private servers to go back to the old DX9 client.

>>101519411
and? It still works. Higher settings, resolutions and graphics are a luxury; not an extream luxury like caviar but more like heated seats in a car. You don't really need them but if you can afford them then they do make considereably difference.

>>101519402
Yeah that's what I'm saying. You buy the PC that fits your budget and upgrade it when it stops performing to your standards. In contrast to saving up for a maxxed out build and then riding it long past its expiration date. I'll concede that it does break down at the top and bottom ends - if your budget was GT1030 tier I would have said to save more.

>>101519385
Lives? Where do you think he's from.
Go watch some of his older videos from 2016-2017.

>>101515514
those seem like pretty high temps even for an OC, does it perform the same at stock clocks?

https://www.tomshardware.com/pc-components/gpus/nvidia-geforce-rtx50-launch-pushed-back-to-early-2025-according-to-prominent-leaker

>>101519504
And what Im trying to say is that the computer parts you buy don't become obsolete as fast you think unless you can't live without luxuries like ultra settings, 1440p+ reoslutions and fps in 3 digits, 50-90fps is enough if you don't want to spend 1000$+ every 5 years.

New Intel without cuckcores - Yay or nay?

>>101519588
didn't you hear the news? new intel is broken.

>>101519588
there isnt anything in computing space where "intel" is needed

>>101519584
Agreed. That's what I was getting at with "your standards". I'm in the same camp as you; bought a 1660Ti in early 2019 and haven't felt the need to upgrade since. Maybe when they make a AAA moviegame worth playing.
What I want you to appreciate is that other people might have different standards and that dismissing them as "luxury" doesn't cut it because /all/ of this is luxury. You don't /need/ 60FPS, or 1080p, or low settings either. You don't need a gaming PC at all.

>>101519570
We're not gonna see new GPUs from either companies until they get a steady supply of GDDR7. Samsung will rev up production soon followed by Micron and Hynix. At best you'll get adoption in late 2025 or early 2026.

>>101519761
I am waiting for glass substrate before I upgrade.

>>101519588
Intel in general is a nay at this point. Wouldn't touch them until they sort out the degradation/oxidation/dust/voltage mess (as no doubt they will in time).
Once that is sorted they'll be an interesting proposition. I expect quite niche though unless they're cheap. Basically, the TurboOC FPSmaxxers don't need them because 8 p-cores with disabled e-cores is plenty for gaming and the Cinebench enthusiasts will get more points with the 16x e-core versions. Leaving only applications which need a lot of strong cores (but not >12). VMs maybe?

File: maxresdefault.jpg (161 KB, 1280x720)

161 KB JPG

>>101519588
The best intel cpu have already been made

>>101519790
that's not 2500K

>>101519797
*2600k

>>101519011
yes but only the lower end skus that havent been tapped out as hard
also memory oc is relevant

>>101514404
the 4090 cant even run maxed settings at 1080p over 50ps averages in 2024
no gpu can max out games in 2024
in 5 years 4090 is gonna be at a slide show maxing out everything

>>101512482
>It broke? Oh well just buy another cheapass car.
stability doesn't matter
what you have on your pc doesn't matter
if shit breaks in mission critical situations it doesn't matter

>>101519790
jewfus was right

>>101518341
agent of chaos, sir

Can a 7600X be air cooled or do I need an AIO? I really can't be doing with an AIO. First time builder.

>>101519974
You can use a Phantom Spirit SE for $35 or even a single tower cooler for $20. No need for an AIO

>>101517487
Hmm, currently XMP does seem to be ok right at the moment. Prime 95 large FFT's, shit is boosting and getting really hot but it's not falling over.

Doesn't make any sense compared to before. I guess I'll just keep hitting it with tests over the next few days and see if it holds up. It just doesn't give you confidence not making any changes as far as I know and getting different results. Weird

File: firefox_22_07_24_17_24_34.png (77 KB, 980x762)

77 KB PNG

https://pcpartpicker.com/list/8jW2Jy

Ordered and acquired the rest, all that's left is waiting until they arrive so I can start building. Thanks for all the help, /pcbg/.

>>101520129
nice build
only thing I would've done differently is to select an asrock or gigabyte board

File: 1721163674178311.jpg (84 KB, 750x600)

84 KB JPG

https://www.youtube.com/watch?v=dNlZqt1yEtc
These benchmark videos usually show very similar framerates, but AMD is running at a higher CPU usage than Intel. Is this a future proof concern for AMD if that much of the CPU is already being used while Intel is much lower, or is the number skewed because the unused percentage on Intel accounts for E-cores which won't be used for gaming anyway?

>>101520222
The number is skewed because the unused percentage on Intel accounts for E-cores which won't be used for gaming anyway

File: 61yDHA2rg4L._SL1500_.jpg (126 KB, 1500x1500)

126 KB JPG

>>101520187
I originally wanted the B650M-HDV (ideally the Steel Legend Wi Fi, but I found it too expensive) but then I spotted a discount for the Tomahawk. I went ahead with it for a few reasons: all the extra ports, ATX size, better codec, and a secondary M.2 slot that goes straight to the CPU and not the chipset (important since I'll put pic related over the SSD so I don't have to worry about its temperature ever again). A bit of a shame I don't get the upgradeability of the Steel Legend with its 5 gen PCIe and M.2 slots, but beggars can't be choosers.

>>101520129
Not bad anon, not bad. However why is everything rather expensive. Especially the ASRock monitor

>>101520129
Never liked buying Corsair parts. Mainly because if you buy them you have to use their software to control it. Your RAM isn't RGB I think so that's fine.
Shit monitor.
Could probably get a cheaper board too depending on where you live.

>>101520259
Yeah the only thing relevant about corsair is their psus

>>101520241
Wrong, they are used for some games

>>101520269
>>101520269
>>101520269

>>101520222
What anon >>101520241 says plus a bit of zen clocks up and down really, really fast so most monitoring tools can't poll fast enough to get an "accurate" reading (which isn't even that accurate when providing a single number for cpu utilisation as not all cores will be working as hard as others).

>>101520248
I've heard only bad things about msi with regards to ram training speed on bootup caused by their use of cheaper pcb's with only 6-layers and firmware issues
I hope yours is fine

>>101520273
Edit: The number is skewed because the unused percentage on Intel accounts for E-cores

>>101520254
I live in Romania and ordered most parts through Amazon.de to poormaxx even harder. The monitor is a total scam, completely agreed; insane how much they jack up the prices. It's not better for other monitors btw.
>>101520259
It was fairly cheap Hynix A-die, can't really complain about the RAM. I'm willing to give the monitor a try based on TFT Central's review, don't have many other alternatives. Yeah, the board was a choice I made for these >>101520248 reasons, it's still quite cheap for a retail Tomahawk.

>>101520285
I've heard it's an issue only the first time and afterwards boot times are fine.

>>101520304
The Dell recommended in the OP is a little cheaper on Dell.com (delivers to Germany), however too late now

>>101520304
It's kind of funny and it's kind of sad that poormaxxing in the EU means cashing out 1.5k euros at the very least.