[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


File: product-jpeg.jpg (219 KB, 1500x1001)
219 KB
219 KB JPG
>It is just the GREATEST cpu in your path
AMDChads I KNEEL
>>
You sure? I get over 40,000 score on mine?
>>
>>101227962
Anon is basically saying the Ford Model T is the greatest car of all time and you're saying you have a Lamborghini in your garage, completely missing the point.
>>
Usecase?
>>
>>101228237
/thread
>>
>>101227914
7800X3D is a meme-tier chip that has specialized use cases. There are superior solutions elsewhere under a larger range of use cases.
Besides, gayming isn't much of a milestone on CPU. Budget-tier silicon can effortless handle gayming crap, but it doesn't stop obsessive min-max types trying to get the best no matter the opportunity cost.
>>
>>101227914
It's the Perf/Watt that boggles my mind with this chip.
It's so power efficient it's insane.
>>
>>101228685
>There are superior solutions elsewhere
such as?
>it doesn't stop obsessive min-max types trying to get the best
people get the best because they want to support companies providing efficient and affordable solutions to their particular use case, your own obsession with larger ranges of use cases tells me that you are no doubt sub-average IQ
i buy AMD because their chips are efficient, and i have a certain respect towards them due to my refined sense of beauty
>>
>>101228771
All Zen4 chips are insanely efficient, if you keep them under 5Ghz (easy to do). 7800X3D's efficiency comes from a happy accident because of thermal limits of V-cache. AMD doesn't boost it as high as non-V cache Zen4 brethren. Intel chips are not that far behind either if you power limit them. It almost like going beyond 5Ghz is a fool's errand that only obsessive "min-max" types are willing to traverse.
>>
File: 1710322789931269.png (154 KB, 438x499)
154 KB
154 KB PNG
>>101228685
>7800X3D is a meme-tier chip that has specialized use cases.
What are those use cases?
>>
>>101228784
>/v/tard kiddie detected
Ultraportable SKUs (tons of them), professional-tier SKUs (Xeons, Threadrippers and EPYC)
Hate to break it you, but gayming isn't that big of a deal in the grand scheme of things in x86 world. It is a showpiece market that has a decent ROI, but it doesn't break or make either Intel or AMD. They just do enough key jingling to keep the vocal minority of "min-max" types engaged and pawn off their "leakly" yields.
>>
>>101228837
> Mundane stuff which doesn't require fancy V-cache
>3D modeling
>Scientific computing
>Decryption/encryption
>AI and machine learning
It is almost if most /v/tards have no real-world exposure and experience. They naively think that their sacred cow is only thing that matters.
>>
i do what i want
>>
>>101228771
So true that's what blew mind too
Add to that it cooled with just air cooler means less hassles. It almost feels like some black magic considering its cost.
>>
>>101228237
anon is saying that the CPU is the bestest, when it clearly isn't good for anything other than running games at 1080p with a 4090 GPU.
>>
>>101228899
>SKUs
nigger what? don't tell me you use that term to reffer to different brands, lmao
>gayming isn't that big of a deal in the grand scheme of things
it's a use case like any other, i buy for my use case
i accept your concession
>>
>>101229177
>brainless v/tard raging this hard because they dare to criticize their sacred cow
I knew most /v/tard were sad creatures, but this is just pathetic
>>
AMD won this round
>>
Not with those drivers.
>>
>>101233050
/thread
>>
>>101227914
>AMD's version Sandy Bridge
It shall be of service to you for many years to come
>>
>>101228771
>>101228809
>>101229053
Holy fuck you "guys" sound like bots
>>
>>101227914
yeah you're right but I don't like AM5 as a platform apparently its slow as fuck to boot, does it ever go away?
>>
>>101235882
literally never had this happen to me it boots in 20 seconds
>>
>>101236309
post or total?
ddr5?
>>
>>101235882
Just update the bios and enable memory context restore.
>>
>>101228809
on zen4 you can lower the vcache or somethings power limits and keep it way cooler. it would throttle on mine without doing it.
>>
>>101235882
Goes away as soon as you open the BIOS and flip the right setting.
>>
>>101228685
>7800X3D
Is the Ryzen (Ryzen) of Gaming CPUs, I implied OP meant Ryzen as a line of CPUs on the whole.
The original Intel 8086, the 6800/6502, the 386/486/the original "Pentium" (586), the Pentiums II/III, the Core2Duo, the K6, the Athlon (XP) and the Ryzen, that's my list of greatest CPUs of all time (the ones combined with slashes I see more of a continuous descendancy).
Some of these are probably biased, in that they were preceded by a shit generation, so Ryzen... sure is great, but *greatest*?
There's also somewhat overlap between digital computing processing and embedded in the beginning.
Guess why.
>>
>>101236758
>The original Intel 8086
Some noteworthy education I recall having watched:
https://www.youtube.com/watch?v=JH2nXMv6yZI
About 50 mins in two parts.
>>
>>101236770
>education
*Edutainment, education was the real thing, lol, derp. Part 2:
https://www.youtube.com/watch?v=LoTx9LQIKEA
>>
>>101236521
>Just update the bio-AAAACCCCCCCCCKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK
>>
>>101228771
>have to lower base & turbo frequency because of 3D cache
>this forces CPU into a better part of its frequency/voltage curve
>ayymdfags think this is magic
All you have to do with other CPUs, both AMD & Intel, is lower the clocks
>>
>>101236528
>link
https://github.com/PrimeO7/How-to-undervolt-AMD-RYZEN-5800X3D-Guide-with-PBO2-Tuner/blob/main/README.md#1-what-are-we-doing

totally helps to keep the turbo boost up there at 4.4 ghz on my 5080x3d. it would jump down to 4.1ghz a lot and it was hitting the throttle limit. temps were like 90C. after lowering mine down more and more it eventaually ended up at 75-80C and the boost held at 4.4ghz
>>
>>101236940
Why are you like this?
>>
>>101237148
what the fuck is this shit man
can't you just set a static fucking voltage and be done with it? or will that shit fry their fragile weak ass cpus?
>>101237162
fuck you nigger
amd agesa updates are 1 step forward 2 step back. gaymd is a bugfest. unstable pile of shit
fuck you
>>
>>101237182
it was that they had volted the v-cache too high and it didn't need it. you can do this and it will do it at windows bootup just go through this once and you'll not have thermal throttling because of the v-cache getting too hot. idk, amd or the mb makers are retards.
>>
5800x3d is better and it doesn't have Microsoft spycrap on it
>>
>>101237192
can't i do this in bios? running weird third party shit interacting with low level shit in OS is a recipe for disaster
>>
>>101237215
>3rd party
fuck you
>>
>>101228685
Holy fuck it's the owner of userbenchmark.
>>
>>101237230
>4) What do we need?
>Debug-cli.7zp from PJVol from Overclocker.net thread
>it's some random russian
>hosted on gdrive
>no source
fuck YOU nigger
>>
>>101228948
>using CPU for AI or machine learning in 2024
Sure thing, bud.
>>
>>101237268
you only need pbotuner, sorry your bios manufacturer locked those settings out for you. all it does it tune the volt limits on the v-cache. code it yourself as you said muh bios is locked. then do it through windows then, yourself. shouldn't be that hard. or run it from some russian hacker, whatever who cares. i dont.
>>
>>101237212
At 4k both are identical.
>>
>>101237182
>amd agesa updates are 1 step forward 2 step back. gaymd is a bugfest. unstable pile of shit
Factually incorrect, works on my machine with 4 sticks AND fast timings. Skill issue.
>>
>>101235882
It's entirely on you for buying a slow booting MSI board since any basic user review will tell you that
>t. Fast booting Asrock am5 user
>>
File: 1719457416312026.jpg (182 KB, 776x1024)
182 KB
182 KB JPG
>>101228685
>7800X3D is a meme-tier chip
Alright then retard if your so smart then go develop your own 3D V-Cache cpu
>>
>>101235244
kek I straight out jumped from Sandy Bridge to Zen4
>>
>>101237589
i thought assrock was ass!!!! dammit im getting butt rock next time
>>
>>101237891
anal boulder is somehow the most competent MB for any AM5 build, it feels wrong but there you have it. stones up your pooper.
>>
>>101237906
anal boulder made me kek I think ive got autisms
>>
Meh not worth it. If you're a richfag just wait and buy ryzen 9000 and rdna 4.

RDNA 3 was the biggest clusterfuck since the 5700xt days with price gouging. I'm not giving AMD my money unless they truly deliver me a 4k 144fps gaming PC for less than $1300 USD( monitor not included.)
>>
>>101238038
>Meh not worth it. If you're a richfag just wait and buy ryzen 9000 and rdna 4.
it is very likely that 9000 will be worse for gaming than 7800x3D
need to wait 9000x3D

>>101227914
I love my 7950x3D
best of both worlds
>>
File: spider-man-1920-1080.png (68 KB, 560x1170)
68 KB
68 KB PNG
>>101238339
>scheduling issues: *blocks your path*
>>
>>101238997
process lasso exists, not the CPUs fault that uneven CPU cores are a fucking nightmare. the same is true for hyperthreading,and the p/e-cores on intel.
>>
>>101239213
as much as I like to minmax hardware through overclocking, having to babysit the scheduling of a cpu by pinning cores per program is a poor man's solution to something that should just work
>>
>>101239477
whats there to babysit? you rightclick the process, assign it to the first CCX and ur done, thats all there is to it.
>>
>>101239488
System informer does this automatically.
>>
>>101239488
that's a really retarded solution though
even intel's p/e cpus don't require such a hack of a solution to just work because they actually engineered a solution that just works
>>
File: file.png (984 KB, 2560x1440)
984 KB
984 KB PNG
>>101239533
>intel's p/e cpus don't require such a hack of a solution
except they do, its the same performance issues in some games just like the cherry picked tombraider one you presented.
>>
File: file.png (723 KB, 2560x1440)
723 KB
723 KB PNG
>>101239533
>>101239567
and HT, all of this shit adds up, the fact that you think it "just works" on one over the other is laughable and tells me you have no idea what the fuck you're talking about. software like process lasso is the only "solution" to these problems as it can automatically assign the affinity to the correct cores, and even switch power plans based on what you're doing. the windows scheduler is terrible for this stuff.
>>
File: file.png (705 KB, 2560x1440)
705 KB
705 KB PNG
>>101239588
>>
>>101239567
>1% slower
that's a tie
7950x3d has a 200mhz advantage on its vcache ccd (5.25ghz) over the 7800x3d (5.05ghz) and yet still falls short of it by more than 5%
>>
>>101239620
yeah? and in multithreaded tasks the 7950X3D is twice as fast, what is your point?
>>
>>101227914
>gayming shit
>greatest
The 5950X/7950X (non 3D) are the best ones. Go be a manbaby about muh 5 FPS back to /v/eddit.
>>
>>101237436
Not if you play games that only use a single CPU thread, i.e. most good games.
>>
>>101228948
none of those benefit from the 3D cache you fucking retard
>>
>>101239631
amd's half baked scheduling bites them in the ass
the 7950x3d is far from the "best of both worlds" unless you're ok with worse gaming performance than the far cheaper 7800x3d and worse productivity performance than the cheaper 13900k/7950x
>>
File: spider-man-1920-1080.png (59 KB, 560x810)
59 KB
59 KB PNG
>>101239670
whatever, process lasso works better than any of the solutions from amd intel or microshit.
>>
>>101239670
>amd's half baked scheduling bites
Intlel's cuck cores still get used all the time in games, there aren't any good schedulers for modern CPUs.
>>
>>101239681
they're so incompetent that they should just opensource their scheduler for fans to fix desu
>>101239683
yeah heterogenous cores suck ass. thankfully single ccx homogenous cored ryzens still exist
>>
>>101227914
>gaming in 1080p
Are you serious? It's 2024 outside. Besides that, even my old 12700K chip is better in compilation time.
>>
>>101239696
they wont for long... amd are implementing their zen "c" cores on upcoming designs so you'll get the pleasure of the same issues that the intel estrogen cores have.
>>
>>101239707
I'll be holding onto the 9800x3d or 10800x3d until the heat death of the universe if they happens
>>
>>101239670
There's a BIOS option to just use the cache cores, most reviews I've seen are retarded and don't seem to know it exists and the same goes for most customers.
>>
File: file.png (36 KB, 550x455)
36 KB
36 KB PNG
>>101239713
not if, but when. they already implemented it on some of the lower end APUs.
>>
>>101239721
that apu is a harvested low-end mobile chip
strix point is 4+8c dual ccx too but its not happening on desktop chiplets just yet
>>
>>101239656
>Mundane stuff which doesn't require fancy V-cache
>>Can't read
/v/tard illiteracy strikes again
>>
thankfully i bought my endgame 10th gen monolithic homo intel cpu that will purely just work for a decade and more. have fun gaymer minmax goys
>>
>b-b-but AMD is faster at all games
>when it's run at 1080p with a 4090
>>do you have a 4090 and game at 1080p?
>umm, no
lol
>>
File: shrugs.png (12 KB, 560x407)
12 KB
12 KB PNG
>>101227914
My 7900X is faster than the 7800X3D in most tasks and costs the same. Hence the 7900X is better.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.