Monday, September 19th 2022

AMD Ryzen 9 7900X CPU-Z Benched, Falls Short of Core i7-12700K in ST, Probably Due to Temperature Throttling

by
btarunr
 Discuss (123 Comments)
An AMD Ryzen 9 7900X 12-core/24-thread processor sample was put through CPU-Z Bench, the internal benchmark of the app. The chip boosted up to 5.20 GHz in the test, and ran at temperatures as high as 86°C, as reported by CPU-Z. It scored 766 points in the single-threaded test, and 11882 points in the multi-threaded one. The single-threaded numbers in particular are interesting. 766 points would put the 7900X behind the Core i7-12700K and its "Golden Cove" P-core by around 3%. In the multi-threaded test, however, the 7900X, with its 11822 points, is in the league of the next-generation Core i7-13700K (8P+8E) processor, which was recently spotted scoring 11877 points with a 6.20 GHz overclock. The 7900X will hence be pushed as a superior alternative to the i7-13700K for productivity and creator tasks, whereas its single-threaded score ensures that it falls behind the i7-13700K in gaming by a fair bit.
Sources: Elchapuzas Informatico, TUM_APISAK (Twitter)

Related News

Add your own comment

123 Comments on AMD Ryzen 9 7900X CPU-Z Benched, Falls Short of Core i7-12700K in ST, Probably Due to Temperature Throttling

#1
Lovec1990
Well it depends on cooling used. But aperently we will either need very good CPU cooler for ZEN 4 or higer Wattage PSU for Raptor Lake
Posted on Reply
#2
PerfectWave
" falls behind the i7-13700K in gaming by a fair bit " wondering how you know this lol
Posted on Reply
#3
Garrus
It has nothing to do with cooling issues.

AMD already showed in their slides that CPU-Z is the least relevent metric for their next gen Zen4 CPUs. Go read those slides from the reveal.

CPU-Z is totally meaningless and doesn't proxy for the average performance at all.
Posted on Reply
#4
phill
I'll wait for the TPU review... These leaks never really say the whole story... :(
Posted on Reply
#6
Garrus


So you can see that CPU-Z literally is the least representative of any benchmark for Zen4 IPC improvements. Don't forget this picture I posted above is IPC, once you add another 10 percent frequency all those improvements are huge.

CPU-Z is the least important metric. Least representative of the average improvement.
Posted on Reply
#7
N3utro
PerfectWave" falls behind the i7-13700K in gaming by a fair bit " wondering how you know this lol
They assume games are using single threaded tasks, which was true for older games but not for the new AAA ones like cyberpunk which makes full use of multi threaded CPUs
Posted on Reply
#8
Garrus
N3utroThey assume games are using single threaded tasks, which was true for older games but not for the new AAA ones like cyberpunk which makes full use of multi threaded CPUs
Games are reliant on single threaded performance, but CPU-Z doesn't represent gaming ST performance. Did the 5800X3D score very high in CPU-Z? Does Zen4? Nope. But AMD already showed some games 30+ percent faster with Zen4 so who cares what CPU-Z says. Mankind Divided for example is 19 percent faster at the fixed 4ghz clocked speed. Probably 30 percent faster at 5.5+Ghz.
Posted on Reply
#9
Valantar
Not all that surprising, given that CPU-Z 1t seems like pretty much a worst case scenario for AM4's architectural improvements, at least according to AMD's own slides.
Posted on Reply
#10
Bomby569
probably a smaller die was a mistake
Posted on Reply
#11
Valantar
GarrusSo you can see that CPU-Z literally is the least representative of any benchmark for Zen4 IPC improvements.
I get what you're saying, but it's not the least representative - that 39% wPrime improvement is much further from the 13% mean than the 1% CPU-Z improvement. But neither are close to representative, obviously.
Posted on Reply
#12
jesdals
Well that single threaded score was not impressive
Posted on Reply
#13
Garrus
ValantarI get what you're saying, but it's not the least representative - that 39% wPrime improvement is much further from the 13% mean than the 1% CPU-Z improvement. But neither are close to representative, obviously.
People don't remember numbers. In a sense of what it communicates, the CPU-Z results suggests no ST improvement which is totally false (and that is how the article incorrectly interprets the results). Once you add clock speed improvements the WPrime implies "large ST improvements" which is closer to reality for Zen4.
Posted on Reply
#14
Valantar
GarrusPeople don't remember numbers. In a sense of what it communicates, the CPU-Z results suggests no ST improvement which is totally false (and that is how the article incorrectly interprets the results). Once you add clock speed improvements the WPrime implies "large ST improvements" which is closer to reality for Zen4.
I'm well aware of that - I said as much myself above. I was simply pointing out that it's inaccurate to say it's 'the least representative' benchmark for Zen4. You could call it a worst case scenario, which it most likely is. wPrime would still be less representative, just in the opposite direction.
Posted on Reply
#15
AusWolf
86 °C... ok, but what cooler was used?
Posted on Reply
#16
dj-electric
Seems like a mild case, maybe thanks to two CCXs
Future AM5 users - do prepare for a completely different heat management landscape.

Also do prepare to rethink if AM4 cooler Z height compatibility was worth the thermal sacrifice of such monstrously thick IHS
Posted on Reply
#17
Oberon
btarunrThe 7900X will hence be pushed as a superior alternative to the i7-13700K for productivity and creator tasks, whereas its single-threaded score ensures that it falls behind the i7-13700K in gaming by a fair bit.
This is an impressively bad take given what we know about this benchmark. A 13700K will score about 840 in ST, putting it less than 10% ahead of the 7900X, which gets essentially NO IPC improvement over Zen 3 in this test. AMD's numbers show an average of 15.3% uplift in IPC over the previous generation games, including two low-single-digit increases. Assuming the 13700K's score IS representative of its general ST performance (which isn't a great bet, actually; the 12700K has a MUCH larger lead over Zen 3 in CPU-Z than it does in games) then at best it will probably tie the 7900X in most games.

Oh, and there's no evidence of thermal throttling either. Scaling the 5900X score by the max boost increase between the two parts gives approximately the correct score.
Posted on Reply
#19
regs
ValantarNot all that surprising, given that CPU-Z 1t seems like pretty much a worst case scenario for AM4's architectural improvements, at least according to AMD's own slides.
wPrime utilizing AVX-512, where supported. Zen 4 implementation is not the best, yet Intel has it switched off for consumer processors.


And per Dolphin Benchmark there is a remark from developer
emulation/comments/x1n2qx/_/imgwx2f
Posted on Reply
#20
Garrus
ValantarI'm well aware of that - I said as much myself above. I was simply pointing out that it's inaccurate to say it's 'the least representative' benchmark for Zen4. You could call it a worst case scenario, which it most likely is. wPrime would still be less representative, just in the opposite direction.
Seems like you're choosing definitions for words that aren't there. A Math to English problem. CPU-Z showed the smallest improvement therefore it is the least representative of the "slower than what you will get" benchmarks. Happy? I'm not really concerned about unrepresentative faster than average results but I get your meaning.:)
Posted on Reply
#21
mama
This is pointless. We will know the truth shortly.
Posted on Reply
#22
Valantar
GarrusSeems like you're choosing definitions for words that aren't there. A Math to English problem. CPU-Z showed the smallest improvement therefore it is the least representative of the "slower than what you will get" benchmarks. Happy? I'm not really concerned about unrepresentative faster than average results but I get your meaning.:)
That's the thing - "representative" doesn't have a direction, it just signifies closeness to/distance from the mean/norm/overall picture, which can go either way. Of course, the use of those unrepresentative numbers will differ massively due to their direction, as will any conclusions drawn from measuring outliers like this. Which just, once again, shows why one should never, ever use a single benchmark workload as the be-all, end-all indicator of anything outside of performance in that specific workload.
Posted on Reply
#23
Bwaze
CPU-Z also showed 5800X3D slower than normal 5800X in single and in multi core. It's very poor at representing anything meaningful.

Since AMD showed almost no IPC improvement in CPU-Z, it kind of matches the frequency increase - 5900X does 677 in single core, 766 of 7900X is 13% higher. Reported maximum boost clock gor 5900X is 4.8 GHz, but it can jump to ovre 5 GHz momentarily. Reported boost clock of 5900X is 5.6 GHz, with theoretical max at 5.85 GHz? That's about 16.5% frequency increase - theoretical, Ryzen processors don't actually perform any tasks at their maximum boost frequencies. So it might be that 7900X's frequency falls even more from running the single core benchmark like CPU-Z? That, or IPC decrease compared to previous generation.
Posted on Reply
#24
Vayra86
Garrus

So you can see that CPU-Z literally is the least representative of any benchmark for Zen4 IPC improvements. Don't forget this picture I posted above is IPC, once you add another 10 percent frequency all those improvements are huge.

CPU-Z is the least important metric. Least representative of the average improvement.
All those improvements are huge? I'm seeing about 9-19% as the most apparent IPC boost in stuff you might actually use someday. Add 10% clockspeed and you're not in 'huge' territory at all.

1% is worst case, but 39% makes about as little sense and the average certainly won't be 20% either.
Posted on Reply
#25
Valantar
Vayra86All those improvements are huge? I'm seeing about 9-19% as the most apparent IPC boost in stuff you might actually use someday. Add 10% clockspeed and you're not in 'huge' territory at all.

1% is worst case, but 39% makes about as little sense and the average certainly won't be 20% either.
A 13% geomean uplift is ... fine? Decent? Something like that. It's well above what Intel tended to deliver in the 2nd-to-10th gen span, but that's about the best you can say about it. It's not unimpressive, but it isn't impressive either. These chips need the clock speed gains to show a major improvement - and 5nm does seem to deliver at that - but this doesn't make me all that hopeful for the future. Hopefully AMD has been working for a few years already on widening their cores (without this guzzling power). That's where Apple gets their massive IPC, and partially where ADL saw its biggest gains from (but also most likely its massive boost power needs).
Posted on Reply
Add your own comment
Nov 21st, 2024 10:19 EST change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts