Richland ES now out in the wild.

[Ferzerp]

I have to say again: Bonjour! I'm a french model!

 

[Zor Prime]

Did you ever hear the joke about the AMD guy who kept telling everyone "IPC does not decrease!"?

That was fail-proof too, and yet...

Marketing Guy Vs. Developer.

 

[sequoia464]

I have to say again: Bonjour! I'm a french model!

I can't be the only one that got this??

 

[Chiropteran]

I can't be the only one that got this??

It's easy to get the joke, but basically pointless- 99.99% of the information in this forum comes "from the internet", so if you simply assume anything on the internet is a lie we might as well close the forum.

 

[Xpage]

Guess I shouldn't have sent him my old T-bird core with the heat spreader sanded down.

 

[Rvenger]

I would be shocked if Steamroller was taped out and running. It would be great news for once but I can't bet on that.

 

[wlee15]

Update

2012-11-17 05:36:16 < BugMaster> Skyler_: nothing really. the only difference from A10-5800K (Piledrive) are Model number (and I am not sure that this model number corresponds to steamroller and not piledrive) and PkgType (FS1r2 vs FM2)
2012-11-17 05:40:51 < Skyler_> it claims to be a kaveri
2012-11-17 05:45:10 < BugMaster> it hase 13h model number so still 10h-1Fh range. from old info steamroller was supposed to be 30-3Fh range (for APUs) but after FX-8300 release (which have 02h model and piledrive CPU was supposed to be 20-2Fh) I am not sure that info was correct
2012-11-17 05:46:20 < Skyler_> That is weird, so maybe it isn't a steamroller
2012-11-17 05:46:31 < Skyler_> It didn't say steamroller, it said kaveri, but kaveri is supposed to be steamroller?
2012-11-17 05:46:49 < BugMaster> and it doesn't have any new instruction support vs trinity
2012-11-17 05:47:13 < BugMaster> yes. if it was named kaveri than it is should be steamroller
2012-11-17 05:49:27 < Skyler_> did steamroller claim new instruction sets?
2012-11-17 05:49:52 < BugMaster> i don't remember so probably no
2012-11-17 05:50:22 < BugMaster> but there was hope for AVX2
2012-11-17 05:55:24 < BugMaster> there is probably way to look if this is really Kaveri or not by looking if GPU part of this APU is GCN or VLIW4
2012-11-17 06:24:40 < Skyler_> BugMaster: the graphics drivers don't work fyi
2012-11-17 06:24:49 < Skyler_> it came with "standard vga" installed
2012-11-17 06:24:53 < Skyler_> and I tried to install amd's but it failed
2012-11-17 06:25:06 < Skyler_> Throttlestop doesn't work;it can't detect the CPU
2012-11-17 06:25:12 < Skyler_> Core Temp thinks it's 1.6ghz, but cpu-z thinks it's 3.2
2012-11-17 06:25:27 < Skyler_> there's a lot of weirdnesses that suggest it's a new chip
2012-11-17 06:25:42 < BugMaster> dunno if gpu-z can work without drivers
2012-11-17 06:28:21 < BugMaster> but yes such weirdnesses are indication that is something new and not simple new revision of trinity
2012-11-17 06:29:53 < Skyler_> which is VLIW and which is GCN?
2012-11-17 06:36:32 < BugMaster> hm. gpu-z doesn't really show this. it only shows only gpu famaly. than dunno how to check it

So it's Kaveri which meshes nicely with the officially released roadmap but not the one leaked recently.

 

[inf64]

If it's Kaveri than it's SR core,simple as that.Also all those problems he lists(GPU as standard VGA,CPU clock uncertain etc.) are typical for a new/unsupported HW with early firmware. Looks promising .
Oh and AMD never stated there will be AVX2 support in SR core but the fact that it doesn't show up in CPUz doesn't mean it's not there.

Thanks for finding this info wlee15 .

 

[CHADBOGA]

Hopefully OBR will soon bring clarity to this matter.

 

[mrmt]

:hmm: seems unlikely since the last official word from AMD was that it hadn't taped out yet, I doubt they went from not being taped out to having working silicon in hand in the span of a month.

IIRC they stated that at Hotchips, in the end of August, so they have sufficient time here. But I do agree with your arguments that if they do have silicon they have an early sample, too buggy to be sent to developers.

If this is true this is very bad news for AMD, as it means AMD is going to compete with Broadwell with 28nm chips, so whatever gains they make from Vishera will be offset by the gains Intel will make with Haswell and Broadwell and die area will be more of an issue here.

Unless they can rework the architecture to get a smaller cache, this chip won't be economically viable for the desktop market, probably for servers only, and even so, what's the point of launching an SKU to sell just a couple of hundred thousand units per quarter?

They will be out of the mainstream notebook market too, as the Kaveri will be as big as ever and won't have the chipset integrations Intel is planning for Haswell/Broadwell.

 

[Idontcare]

If it is the real-deal then that is great news for AMD because it means they have functioning silicon from A0 steppings. Retail release could be much sooner than we were anticipating.

Maybe they learned from the llano/trinity disaster and they don't want to talk about kaveri for fear it will do to trinity inventories what trinity did to llano inventories?

 

[inf64]

Yeah they have a huge pile of Trinity and Llano now,talking about Kaveri so early would not do them any good. Silent and deadly approach is much better

 

[Vesku]

Hope he can figure out whether it's GCN or not.

 

[alyarb]

Kaveri is 28nm bulk hkmg right?

I thought the full-blown Steamrollers for server/workstation are meant for 22 nm... now which ES do you think is more realistic to come out of GloFo considering the date?

 

[inf64]

28nm APU is the best case scenario. 22nm is still a long way off.

 

[mrmt]

Maybe they learned from the llano/trinity disaster and they don't want to talk about kaveri for fear it will do to trinity inventories what trinity did to llano inventories?

Llano swollen inventories weren't caused by consumers waiting for Trinity, but by missed sales targets that left them with huge inventories. If seasonal trends were to be followed this year AMD would have something around 1.5 billion in Q2 and 1.6 billion in Q3, instead they got 1.4 in Q2 and 1.25 in Q3.

If they really taped out Steamroller, I find strange we are not seeing leaked slides with something like "20% "uplift" in <<insert cherry picked benchmark here>> workloads". The leak is either false or Steamroller does not bring something worth talking. Launch would still be a year away, they could use some good news for the future.

 

[ShintaiDK]

Desperate hopes?

It also smells of nothing but bogus.

 

[Maximilian]

I am excited

 

[Idontcare]

Llano swollen inventories weren't caused by consumers waiting for Trinity, but by missed sales targets that left them with huge inventories. If seasonal trends were to be followed this year AMD would have something around 1.5 billion in Q2 and 1.6 billion in Q3, instead they got 1.4 in Q2 and 1.25 in Q3.

If they really taped out Steamroller, I find strange we are not seeing leaked slides with something like "20% "uplift" in <<insert cherry picked benchmark here>> workloads". The leak is either false or Steamroller does not bring something worth talking. Launch would still be a year away, they could use some good news for the future.

Yeah, unfortunately you are making more sense than my own theory, which doesn't mean good things for AMD

Desperate hopes?

It also smells of nothing but bogus.

I was trying to be desperately hopeful but it just doesn't add up as mrmt pointed out, and others, that there is zero chance 22nm samples would be out and about right now...so much more likely this guy who thinks he has a Kaveri probably just has a Kabini. (which would still be cool, but not as cool as the prospects of him having a steamroller ES)

 

[nismotigerwvu]

TBH I'd be more stoked to have a Kabini. I'd love to see what those Jaguar cores can do on that power budget.

 

[mikk]

The next APU will be a Piledriver refresh, unlikely that someone outside from AMD get a Steamroller based APU this year. I guess he didn't realize AMD delayed Kaveri into 2014, so he believes he get one while in reality it's just a Piledriver based APU.

 

[Vesku]

Kabini would be very interesting especially considering he says it wants to run at 3.2GHz. Are kabini's jaguar cores on 28nm TSMC or on GloFo?

 

[Ajay]

The next APU will be a Piledriver refresh, unlikely that someone outside from AMD get a Steamroller based APU this year. I guess he didn't realize AMD delayed Kaveri into 2014, so he believes he get one while in reality it's just a Piledriver based APU.

Trinity is PD. There have been rumors that Richland is also PD + CGN. The only think I can think of is that the XBOX 720 APUs are doing well, in which case it may not have been difficult for AMD to start ramping up for Kaveri. Even if it is 20% faster than Trinity, AMD would be better off not bragging about it so that they can keep shipping Trinity. When it get's closer to release, they can start bragging an dropping the prices on Trinity.

I can't imagine, at this point, that AMD would release Server/workstation SR on anything but 20nm (~2014) then they should be able to get the power envelope down to levels more attractive to their server customers (especially for a drop in socket replacement) and keep the clocks up because that node has an SHP process. Hopefully, by then, AMD can do better than a 15% bump in performance. Rumors put SR IPC at 30-50% > than PD, taking the former as more likely, it could be a respectable CPU. Of course, this and a buck fifty will buy you a coffee.

 

[Ajay]

Desperate hopes?

It also smells of nothing but bogus.

Well, obviously it's half baked if there are no drivers or bios updates. But, I've always thought that AMD needed to keep moving some x86 hardware if they have any hope of sustaining enough cash flow to develop ARM SoCs. For job postings and AMD's own info they are looking to produce pretty much a standard ARM cpu for 2014 and at least a semi-custom SoC by ~2017. If that the case they need to be moving some x86 product till 2017, assuming they get some serious design wins.

Of course by 2017 an ARMv8 semi-custom SoC will be a very late arrival. So I have no idea WTH AMD is thinking, they are making a huge bet on the 64b ARM server market being large enough and yet new enough for them to get a significant marketshare. At this point, a VIA like fate would be a good thing - the market cap on VIA is ~9.7B!

 

[Vesku]

Trinity is PD. There have been rumors that Richland is also PD + CGN. The only think I can think of is that the XBOX 720 APUs are doing well, in which case it may not have been difficult for AMD to start ramping up for Kaveri. Even if it is 20% faster than Trinity, AMD would be better off not bragging about it so that they can keep shipping Trinity. When it get's closer to release, they can start bragging an dropping the prices on Trinity.

I can't imagine, at this point, that AMD would release Server/workstation SR on anything but 20nm (~2014) then they should be able to get the power envelope down to levels more attractive to their server customers (especially for a drop in socket replacement) and keep the clocks up because that node has an SHP process. Hopefully, by then, AMD can do better than a 15% bump in performance. Rumors put SR IPC at 30-50% > than PD, taking the former as more likely, it could be a respectable CPU. Of course, this and a buck fifty will buy you a coffee.

Also 2014 fits better with DDR4, AMD is not in a position to roll a server/workstation DDR3 version and then a DDR4 version shortly thereafter.