Shintai
Supreme [H]ardness
- Joined
- Jul 1, 2016
- Messages
- 5,677
You are not going to get changes to the Windows Scheduler anytime soon. And remember FX? Yes, that was to solve everything for that CPU too.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
Leaked AMD Ryzen Benchmarks?
- Thread starter HardOCP News
- Start date
You are not going to get changes to the Windows Scheduler anytime soon. And remember FX? Yes, that was to solve everything for that CPU too.
You don't seem to understand what is doing what in the computer system so let me take a step back to the basics, following your next replay:
No it is not, they can tweak it to reduce the problem not eliminate it, scheduling is controlled by the programmer even more so in LLAPI's. Graphics tasks only btw, once the driver tells the CPU what it needs, the OS only is there to communicate those needs to CPU. So the amount it can change it, is limited. The CPU does its own scheduling based on the needs of the graphics driver.
If AMD eliminates the problem entirely in Scorpio's APU by tweaking their architecture, which is probably what will happen, then again this 1st iteration of Zen, will not be benefited.
Last edited:
D
Deleted whining member 223597
Guest
Uses 9590, says Kaby Lake "burns too much energy".
If you are talking about heat, the 9590 still dumps more heat into a room even if it shows up as 67C while a Kaby Lake is 80C.
Shintai
Supreme [H]ardness
- Joined
- Jul 1, 2016
- Messages
- 5,677
Uses 9590, says Kaby Lake "burns too much energy".
If you are talking about heat, the 9590 still dumps more heat into a room even if it shows up as 67C while a Kaby Lake is 80C.
He also doesn't know they measure temperature in 2 different ways
Ryzen is NOT Bulldozer, and for you to say that shows how asinine you are. The scheduler can be fixed it doesn't take years of work to fix it. You have proven beyond a doubt that you have no interest in truth.
Sigh...... Do you need links to the windows schedular and what it does? It has nothing to do with how the CPU schedules graphics or application needs, it has to do with window's needs so things don't get screwed up when multitasking.
https://msdn.microsoft.com/en-us/library/windows/desktop/ms685096(v=vs.85).aspx
read.
Then tell me what tells the windows scheduler what is priority and what is to be done at what time.
Then if you want further reading, check out NUMA support, that is where the problem is most likely occurring. Its all DEVELOPER controlled.
So if Windows scheduler is where the problem is, it will happen in all programs not specifically games. That is not where the problem resides!
Please separate Windows scheduler from the CPU instruction scheduler in your mind, they are not one and the same. One is a pass through that checks for its own needs vs the other (where the problem is actually happening) feeds the ALU's of the CPU.
Yeah, I knew those guys were spreading false information to fuel the hype. The funny part is that Lisa Su has just confirmed that Keller is not the lead architect of Zen
For those of you that don't know, next is a photo of the Zen team. Mike is the guy in front. And she is Suzanne Plummer the lead of the team.
I guess the 'geniuses' at SA forums will have a hard time deleting memes like this
http://semiaccurate.com/forums/showpost.php?p=235719&postcount=5
now it is confirmed that Keller is not the lead architect...
Last edited:
EuphoricRage470
[H]ard|Gawd
- Joined
- Apr 15, 2011
- Messages
- 1,105
Yeah, I knew those guys were spreading false information to fuel the hype. The funny part is that Lisa Su has just confirmed that Keller is not the lead architect of Zen
For those of you that don't know, next is a photo of the Zen team. Mike is the guy in front. And she is Suzanne Plummer the lead of the team.
I guess the 'geniuses' at SA forums will have a hard time deleting memes like this
http://semiaccurate.com/forums/showpost.php?p=235719&postcount=5
now it is confirmed that Keller is not the lead architect...
LOL, now can we please tell all the clickbait tech press to stop calling Jim Keller "legendary"? Thanks.
Shintai
Supreme [H]ardness
- Joined
- Jul 1, 2016
- Messages
- 5,677
I remember how people got angry when it was said Keller wasn't the person behind Zen.
That seems a bit of re-writing of history there by Lisa
This is what AMD said at the time of contracting Jim Keller
Basically he was the lead and senior architect-engineer R&D of Ryzen, yeah he did not do all the work but his role is integral and actually pretty much core.
Cheers
This is what AMD said at the time of contracting Jim Keller
But it makes sense that his role would be played down as a) I think was contracted not full time employee b) no longer involved with Ryzen as left last year.
Basically he was the lead and senior architect-engineer R&D of Ryzen, yeah he did not do all the work but his role is integral and actually pretty much core.
Cheers
Last edited:
EuphoricRage470
[H]ard|Gawd
- Joined
- Apr 15, 2011
- Messages
- 1,105
I remember how people got angry when it was said Keller wasn't the person behind Zen.
Next shock: when people find out that Lisa Su doesn't actually help design the chips either.
From PurePC's review: 2014 Haswell-E well ahead of Ryzen in games
Core i7-5960X relative to Ryzen 7 1800X
- Battlefield 1: 17,1% faster
- Crysis 3: 5% faster
- Dishonored 2: 19.5% faster
- Deus Ex (DX11): 21.8% faster
- Fallout 4: 4.4% faster
- Hitman: 28.8% faster
- Rise of the Tomb Raider: 58.4% faster
- Total War: 3.7% slower
- Watch Dogs 2: 15.5% faster
- The Witcher 3: 46.1% faster
Core i7-5960X relative to Ryzen 7 1800X
- Battlefield 1: 17,1% faster
- Crysis 3: 5% faster
- Dishonored 2: 19.5% faster
- Deus Ex (DX11): 21.8% faster
- Fallout 4: 4.4% faster
- Hitman: 28.8% faster
- Rise of the Tomb Raider: 58.4% faster
- Total War: 3.7% slower
- Watch Dogs 2: 15.5% faster
- The Witcher 3: 46.1% faster
That seems a bit of re-writing of history there by Lisa
This is what AMD said at the time of contracting Jim Keller
But it makes sense that his role would be played down as a) I think was contracted not full time employee b) no longer involved with Ryzen as left last year.
Basically he was the lead and senior architect-engineer R&D of Ryzen, yeah he did not do all the work but his role is integral and actually pretty much core.
Cheers
He is very capable not saying that he isn't but to give reason for his capabilities to supersede Intel's 4 or more design teams capabilities, that is just crazy lol.
My point is it contradicts the recent posts regarding Lisa Su and what she says regarding the team
I worked with senior engineers from R&D in a few tech companies, Jim's role fits pretty much in the same way I have seen with similar senior engineers.
My work also involved working heavily with CSO/CTO, who Jim answered to with Papermaster.
In other words I would go with what Papermaster said (who created the structure specifically for Zen not Lisa) that the senior architect is Jim Keller.
Point is recent posts are IMO wrongly downplaying that involvement as a more high level manager, from my experience senior R&D engineers (such as defined by Papermaster for Jim) are heavily involved in the actual engineering project at a fundamental low level.
Cheers
My point is it contradicts the recent posts regarding Lisa Su and what she says regarding the team
I worked with senior engineers from R&D in a few tech companies, Jim's role fits pretty much in the same way I have seen with similar senior engineers.
My work also involved working heavily with CSO/CTO, who Jim answered to with Papermaster.
In other words I would go with what Papermaster said (who created the structure specifically for Zen not Lisa) that the senior architect is Jim Keller.
Point is recent posts are IMO wrongly downplaying that involvement as a more high level manager, from my experience senior R&D engineers (such as defined by Papermaster for Jim) are heavily involved in the actual engineering project at a fundamental low level.
Cheers
Ah yes that I agree with!
EuphoricRage470
[H]ard|Gawd
- Joined
- Apr 15, 2011
- Messages
- 1,105
Ouch.
SighTurtle
[H]ard|Gawd
- Joined
- Jul 29, 2016
- Messages
- 1,410
There were a few people interested in mITX AM4 boards right?
https://www.techpowerup.com/231205/biostar-shows-off-first-mini-itx-socket-am4-motherboard
https://www.techpowerup.com/231205/biostar-shows-off-first-mini-itx-socket-am4-motherboard
That seems a bit of re-writing of history there by Lisa
This is what AMD said at the time of contracting Jim Keller
But it makes sense that his role would be played down as a) I think was contracted not full time employee b) no longer involved with Ryzen as left last year.
Basically he was the lead and senior architect-engineer R&D of Ryzen, yeah he did not do all the work but his role is integral and actually pretty much core.
Cheers
Keller and Feldman were hired by Rory Read for the ambidextrous strategy (your own quote mentions "AMD’s ambidextrous strategy"). Keller was hired to work on K12 and Skybridge projects due to his familiarity with the ARMv8 ISA since he worked at Apple. He leaded a team of x86 engineers and worked in the K12 and in a modification of Jaguar to fit with the Cortex-A57 cores for the goals of the Skybridge project.
Keller was never part of the Zen team. This is myth, like the perennial myth that Keller was the lead architect for K7, K8, and Apple's cyclone. No, he wasn't.
You can check I have been saying for years that Keller is not the lead architect for Zen. Now Lisa Su just confirms what some of us already knew. She is just giving credit to whom deserves credit for their hard work. Zen is the product of the Zen team with Mike Clark being the lead architect and Suzanne Plummer the lead of the team (check photo above).
Shintai
Supreme [H]ardness
- Joined
- Jul 1, 2016
- Messages
- 5,677
When Keller quit, we know when K12 was cancelled for good.
Keller was hired by Papermaster, Papermaster was hired by Rory.
Not answering the rest because that is just as wrong.
I suppose the AMD news brief from AMD I quoted in my earlier post about the employment and role of Keller was a myth as well sigh.
And sorry but considering I work with CSO/CTO and senior R&D engineers, that AMD newsbrief matches my experience with regards to what they said about Keller and from my experience he would be involved low level with Ryzen and pretty much core/responsible for its R&D..
To point it out again.
Papermaster was the one who was organising the structure of this section, Rory gave the go-ahead and money allocation at a high level and not directly involved.
Last edited:
And Papermaster was following the plans of Rory Read...
It is 2017 and still some people don't accept that K7 and K8 were not designed by Keller. I guess by 2037 people will be still pretending that Keller was the lead architect for Zen.
We know he left abruptly. I heard he left after Lisa canceled the K12.
wriggle words and you know it.
By that logic you might as well say CEOs employ everyone and those actually involved in the process do not count because they are only doing what they were authorised by a CEO....
The decision and structure WAS Papermaster, so your OP in response to mine started from the wrong context, and Keller was core to Zen R&D albeit I agree he would not done all the work as it does involve an engineering team (under Keller who was the senior architect for that section).
At the level that mattered it was Papermaster and Keller, Rory had marginal input on the team structure as he is not an engineer. but the frigging CEO....
Last edited:
K12 never really made any sense, it would cause internal competition with their own products, I think it was a back up plan just in case Zen wasn't good enough.
There is some truth on both sides, but guys doesn't matter who hired who? the fall out is here lets see where it goes.
It matters when Juarnga says the information I provided is wrong (even though it was directly from AMD and Papermaster) and he started from wrong context position based upon that.
And like I said, it aligns with my own experience working with CSO/CTO and senior lead engineers for R&D teams.
Same way Papermaster was instrumental in the structure going forward for R&D for SoC and relevent software IP.
http://www.amd.com/en-us/press-releases/Pages/amd-bolsters-engineering-2013jan23.aspx
And separately worth noting it mentions at bottom Jim Keller was employed as chief architect, with one of those employed Meretsky being a colleague of Keller.
Cheers
Last edited:
In reality it started when Lisa Su confirmed in public that Keller wasn't part of the Zen team and the lead architect for Zen is Mike Clark. Then you decided to post #1578 accusing her of "rewriting history".
In reality, she is just making public the information that some of us have had for years and posted in forums...
And the irony is it was Keller who would had made Mike Clark lead architect, Keller would had still been integral to the initial design/arch and then also worked with Mike who he made the lead architect, while Keller is the Chief architect and would take more of a back seat once it was further along but still heavily involved in the engineering aspects.
Cheers
The one aspect I am trying to understand is the senior role between Mike Clark and Suzanne Plummer (also an engineer) who back in 2015 was team leader on Zen and still R&D.
We can only make assumptions of their two roles togerther, which could be very wrong.
Cheers
We can only make assumptions of their two roles togerther, which could be very wrong.
Cheers
Last edited:
Gideon
Supreme [H]ardness
- Joined
- Apr 13, 2006
- Messages
- 4,217
Man I remember locking core affinity on my FX chip to solve issues with how a program was using the chip. After a few windows updates I no longer had to do that and newer programs used it properly and others were patched. Heck I still have the little program I used to use for doing that, have not used it in a long time. If a game hates the SMT in use, just turn it off for that program and it will run fine, from what I have seen.
He supervised the Zen team before he left and that is irrefutable fact. You always try to paint a false picture. You start with a kernel of truth and like a snake always try to twist things.
Master_shake_
Fully [H]
- Joined
- Apr 9, 2012
- Messages
- 17,794
and yet guru3d shows nothing of the sort.
http://www.guru3d.com/articles_pages/amd_ryzen_7_1700x_review,20.html
except for Farcry.
You should check other reviews out too. There is differences in every review it leads me to think that there is a lot of work needs to be on microcode side. Most other reviews had those games not anywhere near that. So idk where the issue is. May be someone just got a shitty board and not udpated their bios? We heard there was a lot of updates rolling around and one reviewer said this other reviewer was getting better results with gigabyte board. So I am just wondering if its software side is just rushed for now, and it will take some time and few bios updates to get all these discrepancies ironed out.
Not denying the fact over all it is around 10% slower in gaming from what I have gathered from reviews.
I cant say what is happening where all these results flipped, but I can give you the heads up on one in particular: Ashes of the Singularity. Stardock fid a major upgrade on the game last week and that broke the AMD optimizations. Brad Wardell of Stardock is working feverishly with AMD to return Ryzens stelar performance in that game. I expect it will completely destroy the 6900k when the fixes are in place in the next 2 weeks.
Broadwell-E is 20% faster per clock in games and that's when you disable HT for both chips, otherwise.. close to to 30% faster according to Hardware.fr. Also 1800X can't overclock, 4% gain vs stock according to ComputerBase. IMHO there is no magic update that will save it from getting spanked by Skylake-X @ 4.5GHz+ later this year in CPU intensive games.
Last edited:
It won't just be gaming, where Skylake - E will be ahead, it will be ahead in everything, if the Intel will be able to clock it like the desktop parts.
lolfail9001
[H]ard|Gawd
- Joined
- May 31, 2016
- Messages
- 1,496
Ryzen sucked in AotS since the very first engineering sample, so that's just making up BS to cover the problem.
And I remember that then some sites did publish the hypothesis that the bad performance was related to some memory issue in the tested engineering sample.
CanardPC got access to some non-official 6C and 4C models. I hope they can provide some benches soon. They got their hand on R5 1600, R5 1400, and R3 1200. Base clocks 3.2GHz and lower, and the single-core turbo is 3.4GHz in the best case. What really surprised me is that they report there is no all-core turbo: "Pas de boost 4C". Most of the discussion is in French, but you can access one English tweet here