Ryzen: Strictly technical

[Justinbaileyman]

This just showed up via Twitter.

Well since AMD and Microshaft cough cough Microsoft wants to hold hands and play dumb like there is nothing wrong with "Ryzens Thread Scheduling"... Then I just wont use Windows at all and go linux full time. Heck I might even just return this Ryzen stuff I juust bought and upgrade my skylake i3 to a Kabylake i7 and pocket several hundred bucks.Nothing wrong with Ryzen in Windows... Yeah right!!

 

[unseenmorbidity]

Well since AMD and Microshaft cough cough Microsoft wants to hold hands and play dumb like there is nothing wrong with "Ryzens Thread Scheduling"... Then I just wont use Windows at all and go linux full time. Heck I might even just return this Ryzen stuff I juust bought and upgrade my skylake i3 to a Kabylake i7 and pocket several hundred bucks.Nothing wrong with Ryzen in Windows... Yeah right!!

I am starting to feel the same way. There isn't really any point in supporting the underdog, if they become just a bad as the monopoly. If I am going to buy from lying sleazebags, then I might as well buy intel and nvidia.

 

[OrangeKhrush]

They still have some interesting options for Zen2, which in part didn't make it into Zen1 due to complexity and available time and other resources.

As I wrote before, they might use Zen1 as a more general purpose core to be used from top to bottom, servers to mobile. But Zen2 might be added to the portfolio as a more specialized core, improving on the remaining weaknesses of Zen1. For example a likely K12 related AMD patent (covering an AArch64 CPU) showed a third AGU. Schedulers, FPRF read ports for FMA, renamer, buffer sizes, SMT partitioning, etc. could still be improved on. Mind you, that Ryzen is where it is with all those trade offs.

I ditto that, I think Ryzen had to lay a base for AMD to build off, that required a stable working uARCH which Ryzen is bar a few teething in issues and there is potential as benches show that it is very well rounded. You get something like a 6900K or 5960X performance with workstation loads and something like a high end i5 or in some instances can compete with baseclock 5960X and in others pips the 2600K/3770K and even the 4770K in gaming. I would put Zen as a perfect working class citizen.

 

[unseenmorbidity]

I ditto that, I think Ryzen had to lay a base for AMD to build off, that required a stable working uARCH which Ryzen is bar a few teething in issues and there is potential as benches show that it is very well rounded. You get something like a 6900K or 5960X performance with workstation loads and something like a high end i5 or in some instances can compete with baseclock 5960X and in others pips the 2600K/3770K and even the 4770K in gaming. I would put Zen as a perfect working class citizen.

I would classify it as a server hand-me-down that doesn't fit.

 

[OrangeKhrush]

I would classify it as a server hand-me-down that doesn't fit.

For sure it has Server like aspects to it, that doesn't mean it doesn't work.

I think it has gone overboard on both sides;

On the one hand you have the anti AMD protaganists that will claim AMD has a broken uARCH and it is doomed until 2020, then you have the AMD protaganists that will claim the schedular is the root of major upgrades into a championchip of all walks. The reality is neither is correct. Ryzen is not broken it was just like a seedling planted and learning to sprout, for the schedular, if we get 5% gains great, 5-10% is about best case scenario, which you never bite the hand that feeds you, that is free performance tweaked showing that Ryzen has adaptable gains unlike Bulldozer that was locked into its microarchitectural grave.

In one week of testing I enjoyed the experience of trying to get the most out of the setup, overclocking is unrewarding at best but there are more ways to skin a cat so they say. All in all not bad for a hand me down server chip.

 

[Justinbaileyman]

I ditto that, I think Ryzen had to lay a base for AMD to build off, that required a stable working uARCH which Ryzen is bar a few teething in issues and there is potential as benches show that it is very well rounded. You get "something like" a 6900K or 5960X performance with workstation loads and something like a high end i5 or in some instances can compete with baseclock 5960X and in others pips the 2600K/3770K and even the 4770K in gaming. I would put Zen as a perfect working class citizen.

Yeah?? "Something like is the key word here!!" I mean come on now.. I dumped my Xeon E5-2670's for this Ryzen build which doesn't even work properly. I could of just upgraded to E5-2690's or V2's for the same price which does in fact work and would perform better.Either that or I could Just upgrade my main rig to a 7700k and be done with it.I dont mind that there are updates and fixes that need to be done.It's all the lying to my face telling me there is nothing wrong at all and ryzen performs just great "No updates needed".That is total bullxxxx when hundreds of thousands, if not millions of users are saying otherwise. Hell yeah there are teething issues for early adopters but AMD and microsoft need to own up to these issue's and fix them.I am not saying all linux distro's work perfect for ryzen but those that do work, "Work". My point is there is something wrong and if linux can fix the issue's on there end why cant Microsoft fix windows??

 

[OrangeKhrush]

Yeah?? "Something like is the key word here!!" I mean come on now.. I dumped my Xeon E5-2670's for this Ryzen build which doesn't even work properly. I could of just upgraded to E5-2690's or V2's for the same price which does in fact work and would perform better.Either that or I could Just upgrade my main rig to a 7700k and be done with it.I dont mind that there are updates and fixes that need to be done.It's all the lying to my face telling me there is nothing wrong at all and ryzen performs just great "No updates needed".That is total bullxxxx when hundreds of thousands, if not millions of users are saying otherwise. Hell yeah there are teething issues for early adopters but AMD and microsoft need to own up to these issue's and fix them.I am not saying all linux distro's work perfect for ryzen but those that do work, "Work". My point is there is something wrong and if linux can fix the issue's on there end why cant Microsoft fix windows??

I have a source inside the PC testing circles, he had access to Ryzen from the first samples to QS samples and he posted two interesting things about Ryzen.

All this talk about waiting for ABC for Ryzen. It's really ridiculous. If you buy something, it should be usable right now. Be it the CPu is at 90% load or 40% load makes zero difference. The frame rate is what counts now and neither I or anyone has the ability to enforce some patch on any developer for AMD Ryzen CPUs.
What is happening with the Ryen CPUs is simply a teething issue that's come about due to being rushed by the powers (the money) that be. AMD missed Q4 2016 and they'd not be able to push it forward to Q2 2017. Right now, AMD already has 2nd and 3rd revision CPUs without all the nonsense and the unfortunate thing is if you bought this round, you'll likely be stuck with with performance. It is something that AMD could have and has mitigate don their end without requiring any aid from developers. The dev kits sent out will help of course, but if you're anyone hoping that an "investment" in Ryzen will help in future, you're sadly mistaken.
Both Skylake-X and the newer revision AMD CPUs are already ahead. AMD's execution was faulty here, had nothing to do with partners and board vendors. As a proof of concept this is fantastic, but it's so much better with the newer CPUs.

Can't say much of anything, but in a couple of month AMD will have for us, CPUs without many of the issues that Ryzen has right now. CPUs that are actually better in various ways as well.
CH6 was meant to be the best board of them all. What went wrong is a minor thing really, in as far as the boards corrupting the BIOS during flashing. That part is entirely ASUS and has nothing to do with AMD. The part related to AMD is that they made several changes during some microcode updates, which effectively locked out the vendors from crucial settings, especially for memory and some power states. The reasons for this from AMD are not important, but suffice to say, being locked out of these areas of the CPU functionality and the other BIOS updating problem happened at the same time on the same board. The CH6 ended up with the highest expectations and the worst combination of issues to be dealing with.

This is not to make an excuse for the motherboard or ASUS, but I'm just highlighting that it was just a perfect storm of problems which led to the perception of having the most frustrating motherboard for the platform. Had it gone the other way, it would have been the best of the lot par none and at some point it actually was with previous pre-release steppings and updates to the uCode of the CPUs.

basically you are actually BETA testing.

 

[lolfail9001]

I have a source inside the PC testing circles, he had access to Ryzen from the first samples to QS samples and he posted two interesting things about Ryzen.

The question is whether it is actually Zenver 2/3 or simply a refresh of Zenver1 with the known issue patched?

 

[OrangeKhrush]

The question is whether it is actually Zenver 2/3 or simply a refresh of Zenver1 with the known issue patched?

I think it is new silicon, AMD while on the one hand it was dastardly underhanded, on the other pure genius, they just rebranded ES's as final release lol and sold them because they knew the market would take them. I have heard things I cannot repeat but the later silicon is very nice.

 

[Magic Hate Ball]

I think it is new silicon, AMD while on the one hand it was dastardly underhanded, on the other pure genius, they just rebranded ES's as final release lol and sold them because they knew the market would take them. I have heard things I cannot repeat but the later silicon is very nice.

Not bad. Hopefully the promise of AM4 motherboards remaining compatible through subsequent Zen refreshes will hold true.

I can re-sell at a loss to my brother and he can upgrade his anemic 4c/8t VM machine with a 8c/16t.

 

[thepaleobiker]

I have a source inside the PC testing circles, he had access to Ryzen from the first samples to QS samples and he posted two interesting things about Ryzen.

That's really fascinating. Is it accurate for me to assume, that by next versions, the source is referring to Ryzen 5 & Ryzen 3? or is he/she referring to Zen v2? (the few months sentence seems to indicate its Ryzen 3/5 but just wanted to confirm)

Thanks for sharing that tidbit - keeps hopes and fires going .

Regards,
Vish

EDIT : ok after I posted this I read additional comments, thanks for clarifying that its indeed Ryzen 3 & 5 that'll be resolved !

 

[OrangeKhrush]

AMD get silicon on availability, so what AMD is selling is first product yields from GF where the process was new and as GF produce new yields the quality in the wafers improves. This normally affects clockspeed, turbos, overclocks and general stability at any clock. Of course AMD may implement higher clockspeeds on the uncore speeds as an example due to better silicon, that would have the effect of reducing latency dramatically.

I would not make the assumption that Zen will automaticlly be 20% faster, but it is plausible that after all tweaks are done and silicon matures that Summit Ridge may be noticeably better and more stable. The kicker is the later you buy the closer to Pinnacle Ridge and the result is that it will not only have new mature yields but also process upgrades. Summit Ridge is good fun but I think AMDs make or break is Pinnacle Ridge, that is where AMD proves to the world the new Zen uARCH is evolving or not.

 

[OrangeKhrush]

That's really fascinating. Is it accurate for me to assume, that by next versions, the source is referring to Ryzen 5 & Ryzen 3? or is he/she referring to Zen v2? (the few months sentence seems to indicate its Ryzen 3/5 but just wanted to confirm)

Thanks for sharing that tidbit - keeps hopes and fires going .

Regards,
Vish

EDIT : ok after I posted this I read additional comments, thanks for clarifying that its indeed Ryzen 3 & 5 that'll be resolved !

It will be silicon maturity across all SR SKU's.

 

[unseenmorbidity]

I think it is new silicon, AMD while on the one hand it was dastardly underhanded, on the other pure genius, they just rebranded ES's as final release lol and sold them because they knew the market would take them. I have heard things I cannot repeat but the later silicon is very nice.

So, this is an es, and I should return it. thx When will the real cpu's arrive? Willthe current boards work, or should I wait for them to be revised as well?

 

[OrangeKhrush]

Not bad. Hopefully the promise of AM4 motherboards remaining compatible through subsequent Zen refreshes will hold true.

I can re-sell at a loss to my brother and he can upgrade his anemic 4c/8t VM machine with a 8c/16t.

depends on his 4C/8T but yeah, brothers are good dumpster divers

I will say I sold my 5960X setup, my old XEON SR2 server and gaming laptop for funds for more important things and I built two identical SFF rigs off 4790's and RX480 8GB cards and I was rather surprised at how well they do. And it makes me wonder really what do you need to get a solid higher end gaming feel.

 

[OrangeKhrush]

So, this is an es, and I should return it. thx When will the real cpu's arrive? Willthe current boards work, or should I wait for them to be revised as well?

The boards will have bios updates regularly so no need for that. Just run your chip for now, but you will be forced to go through the teething pains regardless. You likely have first adoption silicon so you will have what is out the gate silicon on a new uarch, normally that doesn't result in a hassle free experience. Most here who owned first generation core I processors will tell you how shocking 1156 was for a while before they sorted it's issues out.

 

[Justinbaileyman]

depends on his 4C/8T but yeah, brothers are good dumpster divers

I will say I sold my 5960X setup, my old XEON SR2 server and gaming laptop for funds for more important things and I built two identical SFF rigs off 4790's and RX480 8GB cards and I was rather surprised at how well they do. And it makes me wonder really what do you need to get a solid higher end gaming feel.

Noooooo... you sold your SR2 board? Dear god why? That board is Awesome with a capital "A"!!
My Hero VI comes tomorrow to complete my build. I dont know if I should just send all this stuff back and stick with intel or should I keep it and start my build? I guess this is the million dollar question.

 

[leoneazzurro]

The boards will have bios updates regularly so no need for that. Just run your chip for now, but you will be forced to go through the teething pains regardless. You likely have first adoption silicon so you will have what is out the gate silicon on a new uarch, normally that doesn't result in a hassle free experience. Most here who owned first generation core I processors will tell you how shocking 1156 was for a while before they sorted it's issues out.

Or going a little backward, we could talk about Pentium III 1133 Mhz, or the thermal throttling issues on P4 Prescott, or...
(we could also look at the software side, *cough* Windows Me *cough* )

 

[Justinbaileyman]

Or going a little backward, we could talk about Pentium III 1133 Mhz, or the thermal throttling issues on P4 Prescott, or...
(we could also look at the software side, *cough* Windows Me *cough* )

Um pretty sure the first core i series was on LGA1366 not 1156..

 

[OrangeKhrush]

Noooooo... you sold your SR2 board? Dear god why? That board is Awesome with a capital "A"!!
My Hero VI comes tomorrow to complete my build. I dont know if I should just send all this stuff back and stick with intel or should I keep it and start my build? I guess this is the million dollar question.

It is tough but since you got the setup you may as well build it and just find performance parity

 

[OrangeKhrush]

Um pretty sure the first core i series was on LGA1366 not 1156..

lynfield came first, on 1156 which went up to clarksdale only Celerons, i5's and i3's, the socket LGA1366 came out for only i7 Gulftown and Nahelem. 1156 was replaced by 1155 and 1366 replaced by 2011

 

[unseenmorbidity]

It will be silicon maturity across all SR SKU's.

How can silicon maturity alone make a large difference?

 

[Justinbaileyman]

I know that, I had lga1366 i7 920 on launch day and pretty sure that came out before 1156.I could be wrong though..

 

[leoneazzurro]

Um pretty sure the first core i series was on LGA1366 not 1156..

Don't know which Core architecture he was referring... too lazy to check

 

[OrangeKhrush]

How can silicon maturity alone make a large difference?

for stability it can, for the ability to run higher clocks stable it can, for the ability to ru higher frequency through fabric it can. by large difference, will it boost say Tombraider from 112FPS to 140FPS, probably not but it will increase that number enough to be noticeable.