Next generation GPU jump how big really it will be on this time?

[Krit]

Names aside, it doesn't matter too much what Nvidia releases, it should still be a good jump from Pascal.

Let's see my thoughts are that GTX1180 will be ~ 45-50% faster than GTX1080 or ~ 10-15% faster than GTX 1080Ti and of course for marketing purposes it will beat Titan XP by couple of procents.

Remember that GTX1080 was 67% faster than GTX980 and that's a better jump for sure!

 

[las]

I think 1180/2080 will be 10-20% faster than a custom 1080 Ti

 

[Vayra86]

I'm betting on 25-30% performance increase per tier. They cannot jump ahead like they did with Maxwell > Pascal because it would mean a much bigger chip at every price point and it would reduce margins on all of their GPUs with no real reasoning to support that. There is no way the next round of GPUs will be running at 2600-2800mhz. And lacking that clock, they will need many more SMXes to hit 45-50%.

In addition, VRAM / memory wise 25-30% is logical too because they can then make do with similar bus widths for each tier and ride on the higher freq of GDDR6 alone, being @ 256 bit on 1180 and using the 384 bit for 1180ti.

Last but not least this fits precisely in the regular generational bump and thus Nvidia's strategy for Geforce, one that only depends on AMD's moves in terms of timing, and only in a minor sense. Nvidia has no reason to stagnate because the demand for powerful GPUs is higher than ever, but they also have no reason to make a bigger jump because AMD is still chasing last years' performance.

The other option here is that we'll see a very minor performance bump of 15-20% at best as we saw with the Kepler refresh. There are some striking similarities there too and this would provide Nvidia with the largest amount of flexibility (versus Navi, too) ánd maximization of profit. They can then sell their new gen based on the smaller bump and the addition of good performance for new tech like RTX.

 

[Krit]

Vayra from where you getting these 25-30% performance increase ? If GTX1080Ti has the same cuda core count as GTX1180 and is 35% faster than GTX1080. On paper GTX1180 is like GTX1080Ti on steroids because of smaller size 12nm and higher clock speeds with lower power consumption. There in no way that GTX1180 will be slower than GTX1080Ti.

 

[erocker]

Vayra from where you getting these 25-30% performance increase ? If GTX1080Ti has the same cuda core count as GTX1180 and is 35% faster than GTX1080. On paper GTX1180 is like GTX1080Ti on steroids because of smaller size 12nm and higher clock speeds with lower power consumption. There in no way that GTX1180 will be slower than GTX1080Ti.

Unless you work for Nvidia the GTX1180 isn't documented yet.

 

[HD64G]

1180 has got to be faster than 1080Ti. Especially considering the 2nd hand market is gonna be flooded with those. If it isn't, I am waiting for 1180Ti. If I can get 1.5x of 1080 performance I won't mind a high pricetag. I think nvidia knows that in the enthusiast segment the higher the performance, the higher the profit they can make. Refreshing the lineup with marginal performance gain is going to stagnate the sales.

Almost impossible for the performance core to surpass the enthusiast one when on a manufacturing process so close to the one of the previous gen. Can be done only if they sacrifise the efficiency big time. Most possible performance placement for 1180 is between 1080 and 1080Ti, a bit closer to the latter one. My 5 cents.

 

[Vayra86]

Vayra from where you getting these 25-30% performance increase ? If GTX1080Ti has the same cuda core count as GTX1180 and is 35% faster than GTX1080. On paper GTX1180 is like GTX1080Ti on steroids because of smaller size 12nm and higher clock speeds with lower power consumption. There in no way that GTX1180 will be slower than GTX1080Ti.

See you keep looking at 'leaked' and rumored (read: kid's fantasy gone wild) specs and cuda counts, I look at the economical side of releasing a new gen of GPUs and place them in the stack based on how the market is behaving.

The funny thing is that my approach almost never fails - Pascal was the one time I was wrong about a generational jump, and I think we can agree that most people were.

 

[Krit]

Vayra86 i do not understand your point of view! Look at history an you will see that GTX980 was faster than GTX780Ti and GTX1080 was faster than GTX980Ti so why should be GTX1180 slower than GTX1080Ti ? It makes no sense to me and of course for marketing perspective. This time you definitely will be wrong.

 

[las]

"1180/2080" will beat the 1080 Ti no doubt... x70 probably will too (or match it)

 

[Krit]

x70 probably will too (or match it)

On 7nm yes but definitely not on 12nm!

 

[las]

On 7nm yes but definitely not on 12nm!

Depends if arch changes

 

[Vayra86]

Vayra86 i do not understand your point of view! Look at history an you will see that GTX980 was faster than GTX780Ti and GTX1080 was faster than GTX980Ti so why should be GTX1180 slower than GTX1080Ti ? It makes no sense to me and of course for marketing perspective. This time you definitely will be wrong.

- 980 is not that much faster than 780ti, there is at best a 5% difference ie negligible. The 980 runs better when VRAM is pushed hard (delta compression present + 4GB).
- Pascal is the odd one out.

What I'm saying is they will be about equal but the gap between the 1080 and the 1080ti is a little bit larger than 30% so it remains to be seen if it will be precisely equal, effectively what it means is that last gen's performance drops one tier, like I've been saying all the time and as is usual with Nvidia releases.

OR, alternatively, the 1180 will be just a minor refresh like the 680 > 770 was in which case we'll see about 15-20%. This would be the logical next step alongside pushing RTX acceleration on the new gen, it would suit the early GDDR6 that won't be maxed out right away, and it would fit because Nvidia is only competing against itself and the market still eagerly buys up every Pascal card we see.

The only thing that doesn't fit here is saying the 1180 will jump forward like Pascal did. It makes no sense and is not supported by any evidence.

 

[Krit]

The only thing that doesn't fit here is saying the 1180 will jump forward like Pascal did. It makes no sense and is not supported by any evidence.

That will happen only if nvidia will go straight to 7nm but in my opinion that will not happen at this time because manufacture process will make to long time between releases. On average nVIDIA release new graphics cards once in one to two years. Right now we are waiting more than two years already so most likely it will come out in this summer with 12nm process.

 

[R0H1T]

Vayra86 i do not understand your point of view! Look at history an you will see that GTX980 was faster than GTX780Ti and GTX1080 was faster than GTX980Ti so why should be GTX1180 slower than GTX1080Ti ? It makes no sense to me and of course for marketing perspective. This time you definitely will be wrong.

There's a few things different this time, so while I'm not predicting or guesstimating the % increase, there's a good chance that 1180 or 2080 will be closer to 1080Ti than previous gens. Some of the factors include ~

1) There won't a be massive jump in transistor density or frequency coming from 16nm FF, the previous node was 28nm planar so the jump was essentially ~2.5 nodes for Pascal vs Maxwell.
2) The memory bandwidth won't increase as dramatically because GDDR6 isn't actually faster than GDDR5x over the same bus width, 1080 already has GDDR5x so does 1080Ti so the successor to these might have to go with a really wide 512bit mem bus or really high speed GDDR6. The latter isn't coming anytime soon, while the former is highly unlikely due to costs & power.
3) Lastly compute, we've seen Nvidia sacrifice compute for gaming performance in the past. I'm not sure they can afford that in the DX12 & vulkan led era, when games are using more & more compute based tech.

 

[Tatty_Two]

Let's see my thoughts are that GTX1180 will be ~ 45-50% faster than GTX1080 or ~ 10-15% faster than GTX 1080Ti and of course for marketing purposes it will beat Titan XP by couple of procents.

Remember that GTX1080 was 67% faster than GTX980 and that's a better jump for sure!

I don't buy this, why would NVidia offer the consumer more than 20-25% performance increase when they have no competition currently, if it were me and I could drain 45% I wouldn't use it, I would save it up to use when I needed it unless of course there is significant architectural improvements where the improvements are naturally aspirated.

 

[Krit]

Nvidia GeForce GTX 1180/2080 Specs/Performance - What Should We Expect?

 

[StrayKAT]

Considering the games on the horizon (which is what matters to me more than the hardware really), it doesn't seem too drastic an improvement. I'll probably be happy with my hardware for awhile, quality wise. I mostly expect GPUs to push out better 4k framerates and 1440p falling out of favor, but 1080p still being the standard/lowend.

 

[cucker tarlson]

He might be right about next gen really shining in low level api's, async, half precision and ray tracing.
1170 might possibly be just +10% over 1080 in games that don't use those features, and beat 1080Ti in those that do.

 

[Morgoth]

im hoping when the new 1180 comes out the 1080 drops in price on used marked

 

[StrayKAT]

im hoping when the new 1180 comes out the 1080 drops in price on used marked

The new ones finally have become reasonable, along with Vega. Err.. or what they should have been before mining caused a hiccup.

 

[Vario]

The new ones finally have become reasonable, along with Vega. Err.. or what they should have been before mining caused a hiccup.

Right as they are two years old and on the verge of being obsolete. Bad time to buy full price.

 

[eidairaman1]

Lets wait till the cards are officially out on the market.

 

[John Naylor]

Generation to generation is a hard yardstick because the performance increases per tier are not consistent.

As for VRAM, ... much ado about nothing. The reason most folks get fixated on VRAM is not because they haven't researched but because they have relied on tools that don't deliver what they think they deliver. For example, thos most common argument is... "when I use [insert favorite utility here] with my 8GB card, it clearly shows I am using 5 GB, so 4GB is clearly inadequate". The fact is there is no utility that actually measures VRAM usage, what it measures is VRAM "allocation". When you have a 8GB card, the installation routine says "Oh we have 8 GB card so let's **allocate** up to 5GB of VRAM for our usage". When you have a 4GB card, the installation routine says "Oh we have 4 GB card so let's **allocate** up to 2.5GB of VRAM for our usage". In reality, the game may never break 2 GB.

You can compare this to your credit card ... you have a limit of $5,000 with $500 charged caused you bought everybody drinks wint he bar when your team won the Superbowl. Now you apply for a car loan, the "liability" reported to the bank when you apply for a car loan, is not the $500 actually owned, but the $5,000 credit limit.

This has been shown time and time again in actual testing ... alienbabeltech was one of the 1st to really blow this mindset outta the water... they tested like 45 games at multiple resolutions up to 5760 x 1080. There was no significant difference in fps between the 2 GB 770 and the 4GB 770 in most of the games and the 2Gb was often faster. And while there was a difference in a few games, it didn't matter. In these instances, does it really matter if the extra 2 Gb of VRAM gives you 30% more fps when the game is unplayable in either case .? Are you really going to invest the extra $ in a 4GB version of a card to take you from 13 fps to 17 fps ? The most jarring example of this is that max Payne 2 would not install on the 2 GB card at 570 x 1080... so when they installed the 4 GB, they finally expected to show an advantage to the extra 2 GB .... surprise, it didn't happen. As it turned out, after getting in installed w/ the 4 GB card, they swapped it out for the 2 GB card. having fooled ithe install utility's demand for 4 GB, the game played at the same fps (within margin of error), same quality, same user experience.

You can see this at the link below tho the original site is down (not in english but ya can see the data)....

Puget Sound did it with the 690, Guru3D did it with the 960, and Extremetech did it with the 980 Ti

https://www.pugetsystems.com/labs/articles/Video-Card-Performance-2GB-vs-4GB-Memory-154/
http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_960_g1_gaming_4gb_review,12.html
https://www.extremetech.com/gaming/...y-x-faces-off-with-nvidias-gtx-980-ti-titan-x

This is from the last link

First, there’s the fact that out of the fifteen games we tested, only four of could be forced to consume more than the 4GB of RAM. In every case, we had to use high-end settings at 4K to accomplish this. .... While we do see some evidence of a 4GB barrier on AMD cards that the NV hardware does not experience, provoking this problem in current-generation titles required us to use settings that rendered the games unplayable any current GPU.

Yes there are exceptions to this rule... poor console ports for example are notorious for eating up VRAM.

If you look here at TPUs tests on the 3GB and 6GB 1060s you see the same thing. yes, the 1060 6 GB is notably faster than it's 3 GB counterpart (about 6% at 1080p). But let's not forget that the 6 GB is a different card... it has 10% more shaders so it's going to be faster no matter what. Now **if** the VRAM was an issue, we should see the performance advantage gain at 1440p ... it does not.

https://www.techpowerup.com/reviews/MSI/GTX_1060_Gaming_X_3_GB/26.html

This same argument was used on the 970 w/ it's alleged 3.5 Gb deficiency ... yes, if you doid some really weaird things and worked hard to create a problem, uou could..... but not at 1080p or 1440p... and if you bought a 970 to run 2160p, you already made a mistake. Try as they could, various websites tried to duplicate reported issues but were simply unable to do so without 'doing some really weird things" ... and when they could, the 980 exhibited the same behavior.

Getting back to the generational thing ....

The 970 was 39% faster than it's predecessor and just as fast as the 780 Ti @ 1440p
The 980 was 38% faster than it's predecessor
The 960 was 10% faster than it's predecessor
The 980 Ti was 40% faster than it's predecessor

The 1070 was 63% faster than it's predecessor and 14% faster than the 980 Ti @ 1440p
The 1080 was 30% faster than it's predecessor
The 960 was 98% faster than it's predecessor
The 1080 was 73% faster than it's predecessor

Another thing to consider ... The difference between the reference and AIB cards also chnages between generations and model lines.

 

[Krit]

RTX 2080 has higher TDP and less cuda cores than GTX 1080 TI WTF ? AMD truly will ruin good GPU prices and performance doing nothing.

 

[dj-electric]

^This whole thing is a bit funky. There are some RTX 2080 Ti specs there alongside the RTX 2080 name, if you read the wording.
285W TDP belongs to the RTX 2080 Ti, naturally.