• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Polaris architecture – GCN 4.0

I have gone from Titan to 290X to 970 to Titan X to Fury X, so I think I have a fair shout and do love a wind up. Depending on what comes first and what looks like the better option all round, quite possibly my next card will be AMD. By all means though, feel free to think what you like, all good to me :D



Now that's funny :D

:D:D:D:D:D

It was just a laugh but i did throw a small hook in looking for a bite, you never disappointed. All for a bit of fun as it's a rather slow and bad weather day up here.

I don't know why some are so surprised AMD MIGHT be out of the gate faster at a new process node. Ever since 55NM they have been the first to a new process node,and even the shortest time between Nvidia and AMD has been the HD7970 and GTX680 which was a few months. I would be more surprised if AMD and Nvidia launched new cards on the same(or similar) process node at the same time!! :p

+1

Nvidia only seem to release refreshes/new products on the same node around the same time or sooner compared to AMD.
 
Last edited:
Really..... taken from the Anardtech article.



Did you even read my post, I was saying that AMD did it right, that hopefully they have learned from past mistakes and seem to be ahead at the moment. But of course you just have to jump to AMD's defence even though they were being praised and people take the mickey out of DP for NVidia damage control. :rolleyes:

The article didn't disprove what I said and your initial statement was that we couldn't know what was being used for the demo... when the literally only possibility was Polaris. You were praising AMD for something you didn't have right. I know what you were saying, it was still wrong regardless of what your intention was. It's called discussion, or is this the new internet discussion, where if you say something incorrect but it had the right intention it can't be discussed.

Also I replied because of your first point where you said it was standard and fine to use a fake/mockup, I just pointed out the second part because you also said it in the post.

It IS a big deal to mislead people intentionally. I even said this in a post that Kaap and Greg had a go at me about when the first rumours came up that they didn't actually show Pascal, accurate, to which both guys jumped down my throat.

I said, it's not a big deal to show mockups or fakes, it however IS a big deal to show them while pretending they are real and I also said at the time without having watched the conference I don't know if they stated or implied it was real or not and that is where the line is drawn in if it's okay or not. Having now watched the conference my opinion stands as stated a few posts before. They could maybe get off on a technicality of not stating it was Pascal... they heavily implied it was though.


EDIT:- My take for what it's worth is not showing the test system or card means it's a test/not final cooler, meaning it's probably just something zip tied on to work rather than for aesthetics which journalists would leak about how it needs a massive cooler, loud fan or even the opposite, rumours it runs super cool because they put on a tiny test cooler that isn't suitable for long term usage, but would give unrealistic expectations on the final cooler.

This is the other thing, people often show bare dies because especially on an early process you'll have plenty of dead gpus from every wafer which are pretty much perfect for these events. If you made 5 wafers and have 200 working gpus all being tested in various systems and 100 dead ones, which do you parade around and handle in public and on stage without gloves on? That Nvidia couldn't put dead/non working Pascal's on a non working mock up is in fact a bad sign.
 
Last edited:
That isn't the right video

http://www.nvidia.com/object/ces2016.html

That one, from 1:02.50 or so, I actually take it back. He says "this is what it looks like, this IS THE DRIVE PX 2", then goes on to also state, two next generation tegras with pascal gpu and two next generation Pascal GPUs on the back.

Yeah, he flat out lied, worse really is that along with the specs is a picture of the fake rather than a rendered real version.
 
That isn't the right video

http://www.nvidia.com/object/ces2016.html

That one, from 1:02.50 or so, I actually take it back. He says "this is what it looks like, this IS THE DRIVE PX 2", then goes on to also state, two next generation tegras with pascal gpu and two next generation Pascal GPUs on the back.

Yeah, he flat out lied, worse really is that along with the specs is a picture of the fake rather than a rendered real version.

You think we should sue him? He has crushed our dreams now :(
 
:D:D:D:D:D

It was just a laugh but i did throw a small hook in looking for a bite, you never disappointed. All for a bit of fun as it's a rather slow and bad weather day up here.



+1

Nvidia only seem to release refreshes/new products on the same node around the same time or sooner compared to AMD.

It's more the case we always have this discussion every time a new card on s new node is release and people always start debating the fact that AMD will launch first - its been the case since 2007 at least that AMD got into a new node first. This is why I would be more surprised if Nvidia launched first or even within a month of AMD!
 
Nvidia finally made a big concession in bringing the 680gtx long before big Kepler, something that fundamentally screwed them up at 40nm massively and a smaller degree at previous nodes.

It's not just Nvidia's past 7-8 years of struggling on a new process though, Samsung has been significantly ahead of TSMC on their 20nm + finfet nodes for the past two years. They are 6-8 months ahead in every stage, risk production, full production, products to market, yields. The time you can make a profit on lets say a 350mm^2 chip is simply closer at Samsung than TSMC.

As it happens AMD put fewer resources into bringing something up to 28nm while Nvidia put a lot of resources into doing that. AMD were focusing on the next gen/processes while Nvidia focused on the previous node, that will usually put you behind.

Nvidia don't seem to learn, they've done it for the past two nodes on Tegra as well, launch a hugely optimised for a very mature 40nm process while their competitors waited 6 months for the next process to be viable. They had 3 months with an expensive high performance but high power part then over a year behind everyone at 28nm. They did the same at 28nm, a very late K1 while everyone else was a few months away from a 20nm part.

Nvidia for a long time in every segment of the company hasn't handled new processes well for a long time, but they are also with the foundry that multiple customers have moved away from because they are so far behind.
 
You do get that is precisely the point right, they weren't ready to show Pascal.

That is the only point we agree on. Nvidia didn't want to show Pascal at conference showcasing autonomous driving and the partnerships with auto-OEMs.

There is nothing unusual about that in the slightest. AMD also didn't want to show off Polaris, there was no public display of Polaris, only some select NDA partners. Nvidia has done similar with Drive PX2 pascal units, its been shown to select partners under NDA who can't talk about it.


Sure, if there were stacks of Pascal GPUs and public demos on some DX12 games then it would be a good indication that Pascal release is immanent. that will happen at GDC most likely, not at an automotive event.
 
All they had to do was hold up the demo board and say 'This is what the final product may look like, currently we have two Maxwell cards connected for testing and to give you an idea of how it will look. but the final product will come with Pascal boards' queue long talk about performance figures and estimates.

Probably would have been best but at these events companies often don't show the actual product.


PX2 is not supposed to be mainstream availability to Q4 so its not really surprising they only have limited samples 10 months earlier, or that they want to keep some secrets.
 
I have gone from Titan to 290X to 970 to Titan X to Fury X, so I think I have a fair shout and do love a wind up. Depending on what comes first and what looks like the better option all round, quite possibly my next card will be AMD. By all means though, feel free to think what you like, all good to me :D



Now that's funny :D

How long did you have your 290x for before you sold it? Two weeks? Three?

How long did your have your FuryX for, before selling it to Andy? Four weeks?

Why did you get rid of these AMD cards, if you don't care what's in your PC?

What GPU currently sits in your gaming PC? Oh that's right, NVIDIA.

Nothing wrong with that ofc, though pretending you don't care what's in your PC, when your such a big advocate of everything NVIDIA is just silly.

We all know from your countless posts/threads, that you couldn't live without Shadowplay, couldn't live without windowed mode freesync, couldn't live without gameworks effects, plus the other 100 pro NVIDIA things you constantly feel to need to post in AMD threads.
 
I don't know why some are so surprised AMD MIGHT be out of the gate faster at a new process node. Ever since 55NM they have been the first to a new process node,and even the shortest time between Nvidia and AMD has been the HD7970 and GTX680 which was a few months. I would be more surprised if AMD and Nvidia launched new cards on the same(or similar) process node at the same time!! :p

I think it's just certain users are greatly worried/threatened that Polaris may be ahead of Pascal release date wise, and so go instantly on the defence.

Imagine the outcry here if AMD demo'd a fake Polaris? It would be sung from the rooftops for weeks/months to come.
 
How long did you have your 290x for before you sold it? Two weeks? Three?

I think it was 4 weeks but maybe less. Can't remember.

How long did your have your FuryX for, before selling it to Andy? Four weeks?

Yer, 4 weeks I think.

Why did you get rid of these AMD cards, if you don't care what's in your PC?

I had always decided to stick with the fastest available GPU and the Titan X fitted the bill

What GPU currently sits in your gaming PC? Oh that's right, NVIDIA.

A Titan X

Nothing wrong with that ofc, though pretending you don't care what's in your PC, when your such a big advocate of everything NVIDIA is just silly.

No pretending and that is the way it is. Take it or leave it, no skin off my nose and I am an advocate of innovation. You must miss all the pro AMD threads I made and the pro AMD reviews I have done.

We all know from your countless posts/threads, that you couldn't live without Shadowplay, couldn't live without windowed mode freesync, couldn't live without gameworks effects, plus the other 100 pro NVIDIA things you constantly feel to need to post in AMD threads.

Your questions were decent till this little rant. No point me answering really as you seem to have made your mind up.
 
That is the only point we agree on. Nvidia didn't want to show Pascal at conference showcasing autonomous driving and the partnerships with auto-OEMs.

I like how you say it's the only thing we agree on then directly contradict it by saying something different. No, I didn't say Nvidia didn't WANT to show Pascal, I said they weren't ready to, I'm saying they weren't able to. Not able to and able to but choosing not to is not remotely close to the same thing.
 
I think it's just certain users are greatly worried/threatened that Polaris may be ahead of Pascal release date wise, and so go instantly on the defence.

Imagine the outcry here if AMD demo'd a fake Polaris? It would be sung from the rooftops for weeks/months to come.

The only fake company that does such are a nvidia CEO.
You still hear it years from the event and decades later you will still hear that.

Lucky for all Nvidia owners the future is brighter with AMD Polaris :D
The real deal no woodscrew editions ;)
 
I don't get how people can only like one companies GPU's over the other, The only thing that bugs me currently is the locking in of Freesync/Gsync and it is BS, monitors should have both or neither. This is why I'm still sticking to my 1080p 120hz screen for now.

Is there a Fury X replacement due any time soon then? Or is it still going to launch later?
 
I don't get how people can only like one companies GPU's over the other, The only thing that bugs me currently is the locking in of Freesync/Gsync and it is BS, monitors should have both or neither. This is why I'm still sticking to my 1080p 120hz screen for now.

Bottom line your locked in to vendors, but should be mentioned any vendor can implement vrr on FreeSync panels, it's an open standard and any vendor can support FreeSync panels.

Should mention that performance difference between AMD/Nvidia over the years is a matter of a few fps here or there, it's not the end of the world if you do commit to Free/G-Sync.
 
VR is a must for me now and that overrides all of this Free/G-Sync business, so not fussed if I lose G-Sync in truth. I understand it isn't for everyone and most games don't use VR but the games I play and want to play are VR capable.
 
I like how you say it's the only thing we agree on then directly contradict it by saying something different. No, I didn't say Nvidia didn't WANT to show Pascal, I said they weren't ready to, I'm saying they weren't able to. Not able to and able to but choosing not to is not remotely close to the same thing.

They were physically able to, but chose not to. It is a very important difference. They weren't ready to release information to the public, just like AMD weren't ready to.
 
Back
Top Bottom