2070 is a TU106

hedgey · 11 Sep 2018 at 16:32

Yep. This Turding just gets worse and worse.

https://www.techpowerup.com/gpudb/3252/geforce-rtx-2070

Jeps · 11 Sep 2018 at 16:53

Thats a pretty big deal if true so part of me wants to dismiss it as a baseless rumour but given the size of the TU104 die i can actually see this being true.

Ian Evey · 11 Sep 2018 at 22:03

Just have to say meh!

My money is staying in my wallet this round.

4K8KW10 · 11 Sep 2018 at 22:33

445 sq. mm?
So, TU102 754 sq. mm, TU104 500-550 sq. mm and TU106 445 sq. mm..

Really? :confused:

JediFragger · 11 Sep 2018 at 22:35

For context the 1080Ti is 471mm sq.

4K8KW10 · 11 Sep 2018 at 22:41

This would be the first generation where each chip is released in a single corresponding consumer model.
I guess the defective chips will go GeForce, while the fully enabled chips Quadro.

Wrinkly · 12 Sep 2018 at 00:18

hedgey said:
Yep. This Turding just gets worse and worse.

https://www.techpowerup.com/gpudb/3252/geforce-rtx-2070

The GeForce RTX 2070 will be a high-end graphics card by NVIDIA.

I guess Nvidia are writing the blurb.

Simon Belmont · 12 Sep 2018 at 08:02

hedgey said:
Yep. This Turding just gets worse and worse.

https://www.techpowerup.com/gpudb/3252/geforce-rtx-2070

I knew it. It does make sense on why would they put TU104 on 2070. Nvidia spreading the love this gen

D.P. · 12 Sep 2018 at 09:48

The 1070 is a 7.2m transistor 314 mm² die
If is as rumored, the 2070 is a 10.6m transistor 445 mm² die.

That tells you far more than code names. 50% increase in transistors.

KillBoY_UK · 12 Sep 2018 at 11:14

hedgey said:
Yep. This Turding just gets worse and worse.

https://www.techpowerup.com/gpudb/3252/geforce-rtx-2070

already started on the immature names for a product that' s not even been released ? :rolleyes:

Jeps · 12 Sep 2018 at 11:25

D.P. said:
The 1070 is a 7.2m transistor 314 mm² die
If is as rumored, the 2070 is a 10.6m transistor 445 mm² die.

That tells you far more than code names. 50% increase in transistors.

But how much of that 50% increase is taken up by Tensor Cores and RTX Cores? Its gonna be faster then a 1070 but not by the amount that the increase in transistors will have you believe.

bru · 12 Sep 2018 at 13:21

Every body has dismissed these card before we even know anything about performance, just like the Tom Peterson piece on the podcast, he talks for 30 mins about roughly a 40-50% improvement across the board, but the only part that anyone mentions is when he answers a question a 2080 might not always beat the 1080ti or something similar.
40-50% puts even the 2070 ahead of the 1080ti.
Now of course the glory of benchmarking and especially with cherry picking the results it only takes one that shows what you want it too for it to become fact. of course this works both ways.

Taken from here.

https://www.techpowerup.com/gpudb/2840/geforce-gtx-1070

So if the 2070 is in fact the TU106 chip, then NVidia are ahead of AMD with the top 3 chips which put them ahead by a huge margin.

ljt · 12 Sep 2018 at 13:24

With all due respect bru, that graph is for 1080p, a more cpu limited resolution.

The difference between a 1070 and a 1080ti at 1440p and 4k is a much larger gap than 41%.

Jeps · 12 Sep 2018 at 13:27

bru said:
Every body has dismissed these card before we even know anything about performance, just like the Tom Peterson piece on the podcast, he talks for 30 mins about roughly a 40-50% improvement across the board, but the only part that anyone mentions is when he answers a question a 2080 might not always beat the 1080ti or something similar.
.

If they are 40-50% faster across the board, then how can the 2080 possibly not beat the 1080ti in some cases? That very statement contradicts the 40-50% faster argument doesnt it?

bru · 12 Sep 2018 at 13:43

ljt said:
With all due respect bru, that graph is for 1080p, a more cpu limited resolution.

The difference between a 1070 and a 1080ti at 1440p and 4k is a much larger gap than 41%.

Bigger gap but still with in the range.

https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1080_Ti/30.html

bru · 12 Sep 2018 at 13:51

Jeps said:
If they are 40-50% faster across the board, then how can the 2080 possibly not beat the 1080ti in some cases? That very statement contradicts the 40-50% faster argument doesnt it?

Yes I agree it is very confusing, we have to take the across the board comment with certain leeway knowing that their will always be cases where it isn't true. But such is the joy of cherry picking, for example look at the bottom 9 benchmarks in this chart.

https://www.anandtech.com/bench/product/1941?vs=1939

Oh look the 1070 wins all of them, does that make the across the board statement not accurate, well we can all decide that for ourselves, as we all have our own opinions.

NirK · 12 Sep 2018 at 13:52

Problem here is they have no competition any more, so they can do exactly whatever they want. When you used to have some cards which were bargains, like the 970 or various other ones, this was not because Nvidia are being "nice" and want to give you a bargain, it is because AMD had competing cards. Now that AMD are almost completely non competitive, they can do what they want.
Funny because I predicted all of this about 4 months ago and was exactly right on all of it, I should be an industry forecaster or something.

melmac · 12 Sep 2018 at 14:45

bru said:
Every body has dismissed these card before we even know anything about performance, just like the Tom Peterson piece on the podcast, he talks for 30 mins about roughly a 40-50% improvement across the board, but the only part that anyone mentions is when he answers a question a 2080 might not always beat the 1080ti or something similar.

His first reference to performance was this:

"We shared some data that shows a bunch of games, you'll see the perf somewhere between 35-45% better at roughly the same generation. In most cases, if you are on high resolution and not CPU limited, Turing is going to crush it!"

Which is completely at odds with his answer when asked about the presentation slides. Then he claims that the performance over the 1080 will be 50% in most cases.

But after answering that question the interviewer also asked him this direct question, is the 2080 faster than the 1080ti. His reply "I think so. I would expect some cases where 2080 would beat 1080 Ti but I don't have the data in front of me"

I guess people are quoting that last sentence because who really believes he doesn't know the exact performance? And they are so quick to big up any product they are launching. Remember Pascal, in interviews and the launch event, they made it clear that the 1070 was going to be faster than the Titan X. He didn't have missing data for that interview. Surely if it was faster than the 1080ti, they would be doing the same thing now? What's to be gained from hiding the performance?

NirK · 12 Sep 2018 at 16:08

melmac said:
His first reference to performance was this:

"We shared some data that shows a bunch of games, you'll see the perf somewhere between 35-45% better at roughly the same generation. In most cases, if you are on high resolution and not CPU limited, Turing is going to crush it!"

Which is completely at odds with his answer when asked about the presentation slides. Then he claims that the performance over the 1080 will be 50% in most cases.

But after answering that question the interviewer also asked him this direct question, is the 2080 faster than the 1080ti. His reply "I think so. I would expect some cases where 2080 would beat 1080 Ti but I don't have the data in front of me"

I guess people are quoting that last sentence because who really believes he doesn't know the exact performance? And they are so quick to big up any product they are launching. Remember Pascal, in interviews and the launch event, they made it clear that the 1070 was going to be faster than the Titan X. He didn't have missing data for that interview. Surely if it was faster than the 1080ti, they would be doing the same thing now? What's to be gained from hiding the performance?

Basically it will be within about 10% either way, mark my words pal haha.

LePhuronn · 12 Sep 2018 at 16:24

melmac said:
Surely if it was faster than the 1080ti, they would be doing the same thing now? What's to be gained from hiding the performance?

And that's what I found a bit suspect. Not even a footnote or passing comment. Hell, could've even gone like this:

"Hi, I'm Uncle Jensen, and this is your new RTX generation. Of course it's going to be faster than the 10 series, so there's no point in talking about it. You're here for the really good stuff"

I get the RTX technologies should receive all the focus because it's game changing stuff, but to have absolutely zero mention of the normal bits? I dunno, maybe my sarcasm is actually correct, there's little point talking about obvious performance jumps when there's a lot to talk about with RTX.

Still doesn't sit right though...