Polaris not so bright [Hardocp]From ATI to AMD back to ATI? A Journey in Futility

[AnandThenMan]

This guy has lost his marbles.

Maybe, but I think what's more likely is seeing AMD has shut him out he comes out with a flame clicky bait article to make up for lost hits because of no Polaris coverage and reviews.

 

[JDG1980]

Seems like AMD had wanted to sell Polaris 10 for $399 though.

Where was this ever stated?

 

[JDG1980]

I think all rumors point to Fury and Polaris existing together.

No reason to believe that. At most, they might keep the Radeon Pro Duo as a VR-focused card. If they can get even 90% of Fury X performance on Polaris 10, then why would they want to keep manufacturing an outdated chip with expensive HBM?

I don't believe Polaris will match Fury. I am going to guess 5% under 390x 25% less power.

For a full node shrink? That would be an epic fail, on par with Bulldozer. You could lower the clock speeds and voltage on existing Hawaii chips and get ~25% less power usage with a <10% loss of performance. These things are being pushed to the bleeding edge of what the silicon supports in terms of both core clocks and memory. With FinFET and design optimizations, we should be getting Hawaii-class performance at 125W-150W. If lucky, they might be able to pull off Fiji/GM200-class performance by goosing the clocks and voltages, with a TDP of 180W-200W depending on how well the process scales.

 

[selni]

Does anyone actually believe amd won an apple contract with a hot, loud, and power hungry chip? The only thing apple cares about is battery life and decent performance and if amd was significantly behind nvidia in what matters to them they would not win that contract.

This guy has lost his marbles.

Sent from my XT1575 using Tapatalk

Yes? Apple are already using GCN 1.0 for their latest MacBook pros for no obvious reason despite better GPUs from both companies being available. No matter how polaris turns out it's hard to imagine it being behind that.

 

[Mezzanine]

So AMD cut him off, big deal, they have so little market share that's it's not funny. I'm sure Kyle will keep happily reporting on NVIDIA products who are the market and performance leader.

 

[VulgarDisplay]

Yes? Apple are already using GCN 1.0 for their latest MacBook pros for no obvious reason despite better GPUs from both companies being available. No matter how polaris turns out it's hard to imagine it being behind that.

The obvious reason is that the market stagnated on 28nm. It just wasn't possible for them to have a generational leap in efficiency without a new process which they have now.

Sent from my XT1575 using Tapatalk

 

[HiroThreading]

"In the simplest terms AMD has created a product that runs hotter and slower than its competition's new architecture by a potentially significant margin"

ruh roh

http://www.hardocp.com/article/2016/05/27/from_ati_to_amd_back_journey_in_futility

What a poor excuse of a blog.

This man's ego needs to be checked.

 

[trane]

It's a complete non-story. He thought 14nm was magical and Polaris 10 was meant to match 1080. Now even if it ends up at 1070 performance, 150W, $300, it's a disaster by his accounts. Whatever, game is still on.

 

[trane]

I don't mean late to their own plans - but to Nvidia's. It's all relative. Make no mistake, launching after Pascal is not desirable for AMD. Even when they launch first for 5000, 6000, 7000, series they can't secure 50% market share.

AMD were early on 5000, 6000, 7000 series because Nvidia were following on the same process.

A better comparison would be 4000 series. AMD was 2-3 weeks late because they opted for 55nm vs Nvidia's 65nm.

Similarly, AMD have opted for 14nm versus Nvidia's 16nm, and they appear to be 3-4 weeks. Granted, it's a different foundry this time, but Samsung's 14LPP came in after TSMC (they were earlier on LPE, but Polaris is almost certainly LPP). So it's a calculated gamble to go for the better process even if it's a few weeks late.

Let's not forget that 4000 series was AMD's massive resurgence despite the delay.

 

[Elixer]

Wonder if this is tied with that rumour of Polaris not being able to hit its target frequencies(?). I cant imagine HardOCP doing a clickbait journalism either (why would they need to?) and Kyle's put his reputation on the line so definitely he knows something. AMD stuff cannot come fast enough!

Kyle reputation can't be any lower than it is now.
If you looked at the site over the years, you would see pure bias writing at work.
Try to point out the discrepancies, and you will get banned in the forums.
I have no idea why anyone would still throw them clicks, even Steve isn't enough to save that site anymore.
That is all they are now, clickbait, and they know it, and people are falling for it hook, line, and sinker.

 

[tential]

Kyle reputation can't be any lower than it is now.
If you looked at the site over the years, you would see pure bias writing at work.
Try to point out the discrepancies, and you will get banned in the forums.
I have no idea why anyone would still throw them clicks, even Steve isn't enough to save that site anymore.
That is all they are now, clickbait, and they know it, and people are falling for it hook, line, and sinker.

I honestly like his benchmarks. I like the structure and I wish more people would include such results. But his words don't match up with his results a lot of the time and that is troublesome to me.

I want Max 4K testing though from him. I hope he does it when the new Titan comes out and for the new high end cards.

 

[cpo_sharky]

Kyle has been a drunk sellout for many years now. nVidia and Intel have paid him well so he has twisted his site into a sloppy shithole while dragging his forum members with him. Even more sad is he has a small group of 40-50 ballswingers that defend his drunkard rants as if it were theirs.......pathetic.

 

[flopper]

It's a complete non-story. He thought 14nm was magical and Polaris 10 was meant to match 1080. Now even if it ends up at 1070 performance, 150W, $300, it's a disaster by his accounts. Whatever, game is still on.

dont he use google?
we knew that months ago.

 

[Bacon1]

dont he use google?
we knew that months ago.

No he doesn't, he still thinks its impossible to benchmark in DX12. He ignores all the other sites that use PresentMon and Dxtory to capture FPS.

 

[AtenRa]

I honestly like his benchmarks. I like the structure and I wish more people would include such results. But his words don't match up with his results a lot of the time and that is troublesome to me.

I want Max 4K testing though from him. I hope he does it when the new Titan comes out and for the new high end cards.

I also like [H] review methodology and I used to read their reviews but lately I barely visit their site.

 

[cpo_sharky]

I also like [H] review methodology and I used to read their reviews but lately I barely visit their site.

Why?

 

[USER8000]

After mentioning to Kyle that the P10 is rumored to be 230mm^2 and GP104 is 315mm^2 he replied:

Yes, that is a direct quote:
https://hardforum.com/threads/from-...-in-futility-h.1900681/page-8#post-1042319107

After asking for a source, and stating that the A9 doesn't show such a reduction, but only a 8% reduction he said:


As stated, I don't really know anything about fabrication, so you guys can tell me how right/wrong he is.

That explains it!! He thought Polaris 10 was a GP104 competitor due to faulty maths,and so soon to launch when it is obvious that a smaller chip won't do so,he is genuinely surprised. LOL.

 

[3DVagabond]

Interesting.

What HardOCP knows from sources inside and outside the company, AMD has a problem on its hands, as both these products have come up significantly short of where these were supposed to land. But that is OK for AMD, it will simply send Chris Hook out to fall on the sword and tell a story of that was the plan all along....to produce brand new parts much slower than its last high end GPUs. In the simplest terms AMD has created a product that runs hotter and slower than its competition's new architecture by a potentially significant margin. Let the internal finger-pointing begin! The NVIDIA GeForce GTX 1080 and 1070 launch was a surprise to AMD and it simply does not have a harpoon for Moby Dick this time...again. One could go as far as saying that the potential exists for another R600 moment. Now, I am sure AMD will take every step possible to mitigate this gap but the simple reality of the situation as it stands today is that AMD has a loser on its hands and are going to have to pull every trick in the book to spit-shine the turds, and the fact of the matter is pricing is how you do that. All is definitely not fine in Radeon-land these days.

Looks like a Charlie rant with the names reversed.

 

[R0H1T]

Looks like a Charlie rant with the names reversed.

Hotter than the premium 1080 FE, hope this is something worth waiting for :thumbsup:

In the simplest terms AMD has created a product that runs hotter and slower than its competition's new architecture by a potentially significant margin.

 

[vissarix]

What if he is actually right? Polaris at r9 390 level performance and same wattage as a gtx 1070...and maybe runs hotter...thats what he is reffering to...

And about the way he writes...well, just look at charlie demerijan or nick farrell...they try to be different..

 

[selni]

What if he is actually right? Polaris at r9 390 level performance and same wattage as a gtx 1070...and maybe runs hotter...thats what he is reffering to...

And about the way he writes...well, just look at charlie demerijan or nick farrell...they try to be different..

That's not ideal but still a sellable card and represents a significant (but expected given the process) efficiency jump.

Really though there's no point in caring about speculation like this - whether it's sourced from ex-AMD staff or not it's still just rumour and doesn't change very much. If Polaris does end up being disappointing or very late, don't buy it. This article - even if 100% true - doesn't really change that decision in any way.

 

[3DVagabond]

People thinking his article is just pure hate are clearly not thinking logically.

If it turns out later Polaris is amazing product he will lose a lot of credibility and it will also make it impossible to get any AMD product samples in the future.

So from logical point of view he can only lose if it turns out to be false with nothing to gain.

Did you not see his rant on Nano because he wasn't being sampled one? This is his MO when he feels snubbed.

I believe AMD isn't happy with his benchmark suite and they aren't going to treat him like a top media partner until he changes some things.

 

[AtenRa]

Why?

Why i like the methodology or why i dont read [H] anymore ??

 

[sweetusernames]

You know you've messed up when AMD invites Linus, of all people, but not you.

To be honest I don't really read anything on [H] anymore I just go there to look at benchmarks

 

[Yakk]

You know you've messed up when AMD invites Linus, of all people, but not you.

LOL... that's funny, go Linus!