First Steamroller processor core exposure

[Abwx]

I have read that it is between 200MM2 and 210MM2.

Edit!!

This is where I read it:

http://www.hardware.fr/news/13148/computex-apu-kaveri-photo-da-mo.html

HFR has the same spot but with poor visibility , WCCFtech
picture allow to see that the chip is 15 x 18 = 270mm2 ,
that would be 351mm2 with a 32nm process , as much
as 43% bigger than Trinity at equal process.

http://wccftech.com/amd-kaveri-apu-...neration-apu-featuring-steamroller-gcn-cores/

 

[galego]

This chip looks like it s roughly 250mm2 , if it was fabbed
with the same process as Trinity it would be 325mm2 ,
that s 33% bigger , yet Kaveri GPU is only 33% bigger
than Trinitys so it s possible that everything was beefed by 33%
although it makes no doubts that some parts were more
inflated than others.

It is less than 240mm2

 

[Abwx]

It is less than 240mm2

What is your estimation of the size according to the picture that is above , there s a mm graduated rule that should allow more than 10% precision.

 

[Vesku]

Looks like the die in that image is about 240mm2 to me.

 

[Abwx]

That would be only 27% bigger than Trinity at equal node.

Given that the GPU must be 105mm2 the rest of the chip
will benefit only 25mm2 more than a theorical 28nm Trinity
for said sub section.

 

[Idontcare]

That's an absolutely delusional way of phrasing it. AMD sending free samples to reviewers is the same as Intel bribing retailers/OEMs to stop using AMD products? AMD isn't "joining in" on anything. Why don't you get mad at Intel for promising Haswell to be a MAGICAL overclocking beast? That's what their engineer promised on Reddit. So there's an equal amount of misinformation from both companies, and bribing and absolutely unethical anticompetitive practice from only one. I think the bribing and cheating company is the one I'm going to have a problem with.

If you are going to jump into an existing conversation mid-stream and start debating the subject matter then you should at least take the time to bring yourself up to speed on why the conversation has iterated itself to the point in time that you elected to join it.

In this particular case you will kindly take note of the fact that it was the other member (iAMunderDog) who made the assertion that the past is irrelevant (be it AMD's past or Intel's - for which the OEM program you mention would be germane) whereas the present actions by either company are all that matter.

It was to this thought stream (not my thought stream, but that of iAMunderDog's) which I was posting to question why special exceptions were being made (by them, not by me) to excuse one company's behaviors but not the other's.

If you wish to join that conversation and begin critiquing the discussion material then it behooves you to take the time needed to ensure you have the actually correctly identified the context of the ongoing conversation.

So if i make a lot of posts promoting their products i ll
receive a PC.?....

Were you already offered stuff by either AMD or Intel.?.

I wish I was offered stuff by either company But I have never been approached or offered any such freebies (or even stuff at a discount).

However, given the specificity in the targeted individuals by such programs I take this as a positive thing as it speaks to the fact that I am not a fan of any company.

The AMD A-Series Test Drive program targets AMD’s most vocal fans who are active in the component community.

I will admit to being a fan of technology, and on rare occasion I am a fan of a specific product (loved the DEC Alpha 21264PC and FX!32 concept), but sadly (for the various marketing depts out there) I am not amenable to being persuaded to act as a free or cheaply bought mouthpiece on the take.

Every bit of hardware I own was purchased with my own money without any strings attached or business connections involved. I guess in this day and age that makes me "old skool" :\

 

[Vesku]

Same, haven't been approached. I'd be happy to amateur review hardware that I was gifted but I'm relatively independent in my views. Plus I'd feel obligated to point out the source of the hardware even if I didn't meet the FTC guidelines for who has to. Which is why I called out the AMD eval kit someone was reviewing as such in another thread.

 

[SiliconWars]

"Most vocal fans" probably means they have a quick look at post count and the last time they didn't say anything bad about AMD.

 

[guskline]

IDC: Good post. My cpus (3770k, FX8350 and FX8320) were all bought by me. I must confess that the upswing of posters with strident pro AMD, negative Intel has been surprising to me. Perhaps the revelation of this latest "incentive" from AMD explains it. Both AMD and Intel are in a tough business and nothing surprises me. Frankly, I appreciate info on either company's chips. It's just that the poster should be honest about his/her relationship with either company. BTW, my wife and I own some shares of Intel that we inherited from her late father. Trust me, that in no way influences me. The Intel Sandy Bridge 2500k (owned 2) was the first Intel chip I have bought since the old Intel P60. Wanted a 3770k to have the "top chip" for the 1155 socket so I sold both 2500ks.

I was so proud to buy an AMD 386-40 to make my Gateway 2000 Intel 386DX33 run faster that I stayed with AMD a LONG time. I really hope the company can survive and compete with Intel. Owning mostly AMD chips even now (8350,8320, Athlon II 640, Phenom II 965) should show I am hardly "partial" to Intel. BUT when I test my top 8350 against my 3770k (now eclipsed by the 4770k and not even considering the Socket L2011 chips) and state that the 3770k is overall faster, and then get barraged by AMD posters I wonder about the situation. This "incentive" from AMD sure raises an eyebrow.

We win with better and faster cpus. Silly claims like "this chip "crushes" that chip is locker room talk.

Just be honest about whether you are touting one cpu or another because they are faster etc. or because you got a free one from a company who asks you to post positive threads about their company.

 

[frozentundra123456]

IDC: Good post. My cpus (3770k, FX8350 and FX8320) were all bought by me. I must confess that the upswing of posters with strident pro AMD, negative Intel has been surprising to me. Perhaps the revelation of this latest "incentive" from AMD explains it. Both AMD and Intel are in a tough business and nothing surprises me. Frankly, I appreciate info on either company's chips. It's just that the poster should be honest about his/her relationship with either company. BTW, my wife and I own some shares of Intel that we inherited from her late father. Trust me, that in no way influences me. The Intel Sandy Bridge 2500k (owned 2) was the first Intel chip I have bought since the old Intel P60. Wanted a 3770k to have the "top chip" for the 1155 socket so I sold both 2500ks.

I was so proud to buy an AMD 386-40 to make my Gateway 2000 Intel 386DX33 run faster that I stayed with AMD a LONG time. I really hope the company can survive and compete with Intel. Owning mostly AMD chips even now (8350,8320, Athlon II 640, Phenom II 965) should show I am hardly "partial" to Intel. BUT when I test my top 8350 against my 3770k (now eclipsed by the 4770k and not even considering the Socket L2011 chips) and state that the 3770k is overall faster, and then get barraged by AMD posters I wonder about the situation. This "incentive" from AMD sure raises an eyebrow.

We win with better and faster cpus. Silly claims like "this chip "crushes" that chip is locker room talk.

Just be honest about whether you are touting one cpu or another because they are faster etc. or because you got a free one from a company who asks you to post positive threads about their company.

I agree. I dont know what is the motivation of some of the rabid pro-AMD, anti-intel posters that keep popping up on these forums. Ironically though, at least to me, these posts are so riddled with bias and illogic that they do nothing to influence me to buy an AMD product. In fact they have quite the opposite effect of generating ill will toward the company and making me question their products when they have to be supported by such irrational claims.

This is actually unfortunate, because there are scenarios where an AMD product makes sense, but all the blatant fanboyism and intel bashing simply makes me not to want to support AMD at all, and I used to be a fan of them back in the Athlon days.

 

[inf64]

Hmm that's very strange POV, I feel exactly the opposite is true. This is probably the only forum with so many intel shills and fanboys posting with no repercussions(that's why nobody takes it seriously anymore ).

 

[blackened23]

Hmm that's very strange POV, I feel exactly the opposite is true. This is probably the only forum with so many intel shills and fanboys posting with no repercussions(that's why nobody takes it seriously anymore ).

Intel fanboys like me get a free pass. Only AMD fanboys are reviled. Double standard ahoy.

 

[sushiwarrior]

The amount of "OMG AMD SHILLS ERRYWHERE" going on here is ridiculous and absolutely derailing the entire thread. Evidently nobody wants to actually discuss Steamroller, and we'd rather have some kind of circlejerk about shills that don't even exist on forums like this (if you would kindly read your own quotes, you'd notice that they specify it's bloggers and youtube reviewers, not bribing forum members).

Anyways, that sample definitely looks much larger than Trinity or Richland. Maybe it's more beefed up graphics rather than larger SR cores?

 

[inf64]

Oh the cores are def. somewhat larger but iGPU is much more larger than Trinity's one. The size difference in both CPU and iGPU is somewhat compensated by 28nm node process and caches should scale really well with it. Core logic won't scale as well but now we know the 2M Kaveri die (with supposedly full 512SPs and 4MB of total L2 cache) takes approx. ~240mm^2. That's on par with Trinity/Richland today (@32nm). Not bad at all especially when you consider the perf. boost Kaveri should be bringing over Trinity, in both CPU and more so in iGPU department.

 

[Sweepr]

Hmm that's very strange POV, I feel exactly the opposite is true. This is probably the only forum with so many intel shills and fanboys posting with no repercussions(that's why nobody takes it seriously anymore ).

Just because people reply to your overly optimistic AMD predictions/overly pessimistic Intel predictions here (and not in SemiAccurate and other AMD fanboy forums) it doenst mean AnandTech is full of Intel shills. I'll have to admit K10 spanking Conroe by 20-30% clock per clock, Bulldozer running circles around i7 2600K was pretty fun though.

@ Thread: Very interesting post by pc.watch: http://pc.watch.impress.co.jp/docs/column/kaigai/20130704_606220.html?ref=rss

 

[inf64]

It's one thing to predict and be wrong, it's other to be a intel shill/fanboy who derails AMD threads, personally attacks/insults users who disagree, accuse others for things he does himself and all that with no repercussions. Sorry but this happens only here and nowhere else. That's why it's funny and sad at the same time.

edit:
This is one example how SR topic became "AMD shill/fanboy topic". People who came here to read about SR core will find very little on-topic information I'm afraid and we know why is that. Paranoia is strange illness.

edit #2:
Nice ninja edit Sweeper (in order to seem you are on topic at least a little bit ), it's been already posted btw.

 

[Ajay]

IMHO, this is going way OT, maybe a new thread should be spun off?

 

[Abwx]

I'll have to admit K10 spanking Conroe by 20-30% clock per clock, Bulldozer running circles around i7 2600K was pretty fun though.

Not the thread , but worth mentioning in some case it was more
according to intel....

http://forums.anandtech.com/showpost.php?p=35226491&postcount=85

 

[inf64]

Yep Abwx, it's up there with unbelievable OCing of Haswell(Francois where are thou?) and new leaps in IPC by it too. It's marketing talk and we should get used to filter it.

 

[Abwx]

Oh the cores are def. somewhat larger but iGPU is much more larger than Trinity's one. The size difference in both CPU and iGPU is somewhat compensated by 28nm node process and caches should scale really well with it. Core logic won't scale as well but now we know the 2M Kaveri die (with supposedly full 512SPs and 4MB of total L2 cache) takes approx. ~240mm^2. That's on par with Trinity/Richland today (@32nm). Not bad at all especially when you consider the perf. boost Kaveri should be bringing over Trinity, in both CPU and more so in iGPU department.

Excluding the GPU there s 25mm2 left to improve the IMC , I/O and cores ,
wich are 22% bigger than Trinitys assuming all blocks were dedicated
the same ratio , wich highly unlikely imo.

 

[inf64]

Well the die is definitely around 238-240mm^2 as now we have a nice picture with a ruler next to it. We also are not sure if it's GF or TSMC who produces these chips.

 

[Abwx]

Yep Abwx, it's up there with unbelievable OCing of Haswell(Francois where are thou?)

They didnt lie , though , it was actualy...unbelievable..

 

[Abwx]

Well the die is definitely around 238-240mm^2 as now we have a nice picture with a ruler next to it. We also are not sure if it's GF or TSMC who produces these chips.

Most of their APUs and CPUs are still fabbed at GF
and they re WSA bound so they have no choice
than staying with them , hope the process will
retain the SOI waffers.

 

[inf64]

Most of their APUs and CPUs are still fabbed at GF
and they re WSA bound so they have no choice
than staying with them , hope the process will
retain the SOI waffers.

I'm not so sure it's GF. With all the delays and low yield reports I wouldn't be surprised if it was somehow made by TSMC.

 

[CHADBOGA]

Just because people reply to your overly optimistic AMD predictions/overly pessimistic Intel predictions here (and not in SemiAccurate and other AMD fanboy forums) it doenst mean AnandTech is full of Intel shills. I'll have to admit K10 spanking Conroe by 20-30% clock per clock, Bulldozer running circles around i7 2600K was pretty fun though.

Some people are just incapable of learning from their mistakes.

Rather than be cautious about their wishful thinking clouding their judgement, they double down on their silliness and predict that Bulldozer would be the fastest desktop processor.