-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD Zen 3 (5000 Series), rumored 17% IPC gain.
- Thread starter AStaley
- Start date
- Status
It's not 'worrying' as his results make no logical sense. A 5900x gaming 30% slower than a 3900x despite being 20% more powerful with lower latency? It's clearly a problem on his side or with the game.
Last edited:
Soldato
- Joined
- 26 Dec 2005
- Posts
- 16,138
- Location
- Paisley
I was all set to swapover to one of the new chips, i was pretty sad to see that 4k benchmarks there was basically no difference over the 3900/3950x, I presume is that because at that res everything is totally gpu limited? Ill wait and see benchies of it running compared to a 3900/3950 when the new AMD cards dropped before deciding whether its a worthwhile upgrade or not.
Very disappointing launch in terms of reviews, I feel like we had more info from jumper then from all of those 5950X reviews...
Still no proper comparison 5800x or 5900X, no tweaking, no testing C patch BIOSES (they were available earlier for asus at least).
Still no proper comparison 5800x or 5900X, no tweaking, no testing C patch BIOSES (they were available earlier for asus at least).
Caporegime
- Joined
- 8 Jan 2004
- Posts
- 32,613
- Location
- Rutland
Yup at 4K the GPU bottleneck stops much difference being apparent between CPUs.
Every clock starts with a tick!It’s not just Far Cry, the Shadow of the Tomb Raider results are equally as bad. 2/5 games tested showing unexplained results.It's only Far Cry New Dawn, not a game to worry about
Disagree. It is worrying. When disabling 1 CCD his FPS (in SotTR) matched the 10900k, but with both enabled it was crippled, if he doesn't have a faulty chip the new CCD/CCX topology used for the 5900X could be introducing problems under certain situations. He's an experienced benchmarker and can't explain the results, so it'll be interesting to see if others replicate it. The translated article is a bit hard to read, but he seems to suggest it's happening under certain (non-game) workstation loads too.
https://www.igorslab.de/en/amd-ryze...-wird-intels-10-generation-jetzt-obsolet-2/6/
Last edited:
If all the other reviewers are having different results (that agree with each other) then there's something wrong with Igor's Lab results. Maybe he's not using a new BIOS he's supposed to be using.
He said he's tried multiple motherboards with every BIOS provided to him.
Just noticed though, the other reviews (Hexus, Techpowerup) are using a 2080 Ti so the FPS doesn't actually hit the frames at which point the massive disparity between the 5900X / 10900K becomes apparent because of the GPU bottleneck, whereas Igor is using a 3090.
So is Hardware Unboxed and they don't show any of the issues Igor does.
But Hardware Unboxed is using a 5950X which has a different CCX/CCD layout to the 5900X unless I missed another review they've done? Are there any other reviews of the 5900X with a 3090?
What is going on there? Every other review I've seen shows the 5900X matching the 10900k in all games, yet he's showing a few games the 5900X gets absolutely trounced by everything (even a 3600X) unless you completely cripple it by disabling one CCD?
A bit worrying...
simethibg wrong there. Linus, hardware unboxed, gamers nexus, hardware canucks, optimum tech - I've looked at all these reviews today and they show the 5900x beating intel in those games especially tomb raider
Eh? They're both dual chiplet designs, the only difference is the 5900X has 2 cores disabled in each chiplet to be 12 cores. It's the same CCD layout and the same CCX layout, just fewer cores in the CCX.
None of those tested the same situation where Igor hit the problems unless I'm mistaken. 5900X, 3090, 1440p/4K ultra, Shadow of Tomb Raider or Far Cry New Dawn (but no doubt there will be others). Hardware Unboxed tested the 5950X, Gamer Nexus only tested 1080p, hardware canucks/optimum tech didn't test those games.
It affected ONLY the 5900x with 2x CCD's enabled, same problem on mutiple otherboards, and no other reviews that I have seen have had this issue at 4k in games. Seems like a defect or some other abberation.
@PaulM
Just flicked through JayZTwoCents and he/they tested both the 5900X and 5950X with the MSI 3090 Gaming X Trio. Both CPUs matched or outperformed the 10900K. And SotTR was in their list at 1080 and 1440p.
Dude, Igor has a duff CPU.
Just flicked through JayZTwoCents and he/they tested both the 5900X and 5950X with the MSI 3090 Gaming X Trio. Both CPUs matched or outperformed the 10900K. And SotTR was in their list at 1080 and 1440p.
Dude, Igor has a duff CPU.
- Status