• 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.

Poll: 390X: How fast do you think it will be.

How fast will the 390X be?

  • Faster than the Titan-X @ 1080P - 1440P - 4K

    Votes: 59 22.1%

  • Faster than the Titan-X @ 4K only

    Votes: 15 5.6%

  • Within +/- ~5% of a Titan-X @ 1080P - 1440P - 4K

    Votes: 70 26.2%

  • Within +/- ~5% of a Titan-X @ 4K only

    Votes: 16 6.0%

  • Faster than the GTX 980 @ 1080P - 1440P - 4K

    Votes: 60 22.5%

  • Faster than the GTX 980 @ 4K only

    Votes: 10 3.7%

  • Within +/- ~5% of a GTX 980 @ 1080P - 1440P - 4K

    Votes: 17 6.4%

  • Within +/- ~5% of a GTX 980 @ 4K only

    Votes: 4 1.5%

  • 94% or less the performance of a GTX 980.

    Votes: 16 6.0%
  • Total voters
    267
Still lot's of potential for disappointment.
Lack of die shrink, late to the party, performance could be lacklustre.
Peeps already bought into nvidia won't really be tempted to swap over to the new amds if there's no compelling performance increase.
And why have more threads like this being made when we have 110 pages of pure garbish in the rumoured r9 390 thread discussing the same thing?
 
humbug said:
For sure but not unrealistic.

A 1796 Shader GCN 1.2 @ 920Mhz 256Bit 190 Watt 285 is nearly as quick as its 2048 Shader 1000Mhz 384Bit 220 Watt big brother, sometimes a little quicker, it wouldn't take a lot of tweaking to get a 2048 Shader 512Bit 4GB Tonga around 290X performance at 180 Watts.

And it would have to be to compete with the GTX 970, or they might as well not bother.

http://www.techpowerup.com/reviews/Sapphire/R9_285_Dual-X_OC/25.html
Click to expand...

true i guess, but i get skeptical with paper based stats as they more often than not turn out to not perform as they do on paper usually worse but given the evidence of how the 285 performs it would back up that the 380x should easily be faster than a 290x at a lower power consumption. Which would put it past the 970 but right on par with a 980. Then the 390x which will probably be faster than a 980 but right on the doors of a titan x?

Thats how it would look from paper based stats right but obviously with the 390x being more speculation than anything.
 
Davedree said:
Still lot's of potential for disappointment.
Lack of die shrink, late to the party, performance could be lacklustre.
Peeps already bought into nvidia won't really be tempted to swap over to the new amds if there's no compelling performance increase.
And why have more threads like this being made when we have 110 pages of pure garbish in the rumoured r9 390 thread discussing the same thing?
Click to expand...

lack of die shrink -- not dependent on AMD
late to the party -- again, it might not be 100% dependent on AMD, think of HBM. This is the FIRST implementation of a new technology that is more complicated than the previous. Also large availability of HBM might have played a role in 390 series being late.
as for performance, at the worst possible case 390X should at least be as good as 980, anything less is unfathomable.
 
EXxx said:
I'm not sure we really want the 390X to have the same performance as Titan X or more.
i do hope 390X is a 5-10% slower than TitanX and 390 is around 980 level.
what i do really hope is price wars. If 390X is better than Titan X all is nice and toasty but that will enable AMD to charge "an arm and a leg" for it.
I'd prefer it being slower, but at a interesting price.
Click to expand...

EXxx said:
lack of die shrink -- not dependent on AMD
late to the party -- again, it might not be 100% dependent on AMD, think of HBM. This is the FIRST implementation of a new technology that is more complicated than the previous. Also large availability of HBM might have played a role in 390 series being late.
as for performance, at the worst possible case 390X should at least be as good as 980, anything less is unfathomable.
Click to expand...

Forget the GTX 980 both the 390P and 390X will thrash it with little effort at high resolution in particular.
 
unfortunately I'm not past 1080p yet (2 * 24" IPS's) so as much as it's hyped, at the moment I'm not interested in 4k. 1440p is the only way to go for me atm, and at 1440p 980 still stands stronk.
If it were my wettest dream :
380X would be about 970 performance +5% so it has just the slightest advantage
390P would be about 980 performance +5% so it has just the slightest advantage
as for 390X ... trading blows with Titan X at lower price.

@1080p you still need the 295x2, or CF/SLI to run maxed at >60 some games (DA:I and the incoming TW3 will also kneel most gpu's with modded textures)

@ 4k i think 290X showed it's power so there is no reason to imply that 390,390X will show how great they really are

5 weeks? gosh it's slow
 
EXxx said:
lack of die shrink -- not dependent on AMD
late to the party -- again, it might not be 100% dependent on AMD, think of HBM. This is the FIRST implementation of a new technology that is more complicated than the previous. Also large availability of HBM might have played a role in 390 series being late.
as for performance, at the worst possible case 390X should at least be as good as 980, anything less is unfathomable.
Click to expand...

Lack of die shrink most def dependent on AMD, not just Tsmc.
Late to the party most certainly dependent on AMD.
980 performance as a minimum for the rumoured 390x would be an absolute disaster, as HBM would not be required for a 20% imcrease over a 290x.
I can't predict the performance, but lets hope AMD reworked a totally new gcn, as opposed to doubling up the tonga configs of 3584/4096.
 
Davedree said:
Lack of die shrink most def dependent on AMD, not just Tsmc.
Late to the party most certainly dependent on AMD.
980 performance as a minimum for the rumoured 390x would be an absolute disaster, as HBM would not be required for a 20% imcrease over a 290x.
I can't predict the performance, but lets hope AMD reworked a totally new gcn, as opposed to doubling up the tonga configs of 3584/4096.
Click to expand...

They don't need to shrink dies, Hawaii is 40% more GPU on Shaders alone vs Tahiti, it also has a larger IMC, 2x the ROP's, 4x the ACE units

Tahiti is 365mm2, by all accounts Hawaii should be about 550mm2 (at least) like NV GK110, its not, its 438mm2, well over 100mm2 smaller than it should be as a Tahiti architecture.

The reason its so small in relative terms is because they redesigned it and use much denser transistor packing, there have been rumours of AMD packing even denser transistors with the 300 series, but even without that with 4096 Shaders it would be ~635mm2, the Titan-X is 601mm2.
 
Last edited:
humbug said:
They don't need to shrink dies, Hawaii is 40% more GPU on Shaders alone, it also has a larger IMC, 2x the ROP's, 4x the ACE units

Tahiti is 365mm2, by all accounts Hawaii should be about 550mm2 (at least) like NV GK110, its not, its 438mm2 well over 100mm2 smaller that it should be as a Tahiti architecture.

The reason its so small in relative terms is because they redesigned it and use much denser transistor packing, there have been rumours of AMD packing even denser transistors with the 300 series, but even without that with 4096 Shaders it would be 635mm2, the Titan-X is 601mm2.
Click to expand...

If they had a die shrink 2048 tonga would have been 200mm with power draw of around 100-130 real watts. I'm aware of the denser lithography, it's also similar to the carrizo using gf28shp to rescale denser at the sacrifice of clockspeed over kaveri.

A die shrink was what AMd were relying on, this is where NVidia excelled AMD by re-engineering on an old process without sacrificing clockspeed. As AMd have less efficient shader throughput they require a larger die to gain parity/compete with nvidia.
It is certain that a die shrink is needed far more for Amd than NVidia, unless AMD can either rework gcn or create a new architecture
 
Davedree said:
If they had a die shrink 2048 tonga would have been 200mm with power draw of around 100-130 real watts. I'm aware of the denser lithography, it's also similar to the carrizo using gf28shp to rescale it's self denser at the sacrifice of clockspeed over kaveri.

A die shrink was what AMd were relying on, this is where NVidia excelled AMD by re-engineering on an old process without sacrificing clockspeed. As AMd have less efficient shader throughput they require a larger die to gain parity/compete with nvidia.
It is certain that a die shrink is needed far more for Amd than NVidia, unless AMD can either rework gcn or create a new architecture
Click to expand...
I don't know why you are bringing CPU's on different manufacturing nodes even relative to eachother, your just comparing Apples to Oranges, its not relevant all it does is add an irrelevant curve ball to confuse the argument.

Its actually the other way around, AMD have packed more GPU on the same size die than Nvidia (certainly Kepler), the clock rate differences are nothing to do with with those dies or the transistors on them.
Nvidia have a new firmware power delivery system with Maxwell, its able to switch between high and low power states at very high frequency resulting in low average power consumption.
But its only able to do that in a (goldy locks zone), overclocking, even some factory overclocked 980's run at 250 Watts average power draw, just like a 780TI, tho the 980 having higher clock rates.
There isn't actually that much architectural difference between Kepler and Maxwell, aside from Maxwell having better Texture compression technology (just like Tonga)
Its able to clock higher than AMD's current line of GPU's, and Kepler purely because of its power efficient firmware, that allows it to run higher clocks.

A 1500Mhz 438mm2 290X would be faster than a 1500Mhz 398mm2 980.

There is nothing to stop AMD from introducing their own power efficient systems.

Pre- 290X it was said by some in here that such a GPU could not exists on the AMD side, to match or beat a GTX Titan because it would have to be to big, such was the confidence of these people.

In the end AMD had a GTX Titan of their own and it was much smaller than the Titan. Shock!
 
Last edited:
humbug said:
I don't know why you are bringing CPU's on different manufacturing nodes even relative to eachother, your just comparing Apples to Oranges, its not relevant all it does is add an irrelevant curve ball to confuse the argument.

Its actually the other way around, AMD have packed more GPU on the same size die than Nvidia (certainly Kepler), the clock rate differences are nothing to do with with those dies or the transistors on them.
Nvidia have a new firmware power delivery system with Maxwell, its able to switch between high and low power states at very high frequency resulting in low average power consumption.
But its only able to do that in a (goldy locks zone), overclocking, even some factory overclocked 980's run at 250 Watts average power draw, just like a 780TI, tho the 980 having higher clock rates.
There isn't actually that much architectural difference between Kepler and Maxwell, aside from Maxwell having better Texture compression technology (just like Tonga)
Its able to clock higher than AMD's current line of GPU's, and Kepler purely because of its power efficient firmware, that allows it to run higher clocks.

A 1500Mhz 438mm2 290X would be faster than a 1500Mhz 398mm2 980.

There is nothing to stop AMD from introducing their own power efficient systems.

Pre- 290X it was said by some in here that such a GPU could not exists on the AMD side, to match or beat a GTX Titan because it would have to be to big, such was the confidence of these people.

In the end AMD had a GTX Titan of their own and was much smaller that the Titan.
Click to expand...

Lol behave, it's all relevant. Amd rescaling kaveri into carrizo both using the same.process 28shp, is no different to tahiti 28nm rescaling to your denser hawaii architecture using the same 28nm and then concluding it's 100mm smaller than it should be. So don't know why you are claiming a curve ball. It's getting more transistor count into a smaller die than without rescaling and having a larger die.
But you sacrifice clock speed to do it, so you have to have good efficiency in the front end of the gpu.
You will never see a hawaii clock at 1300mhz comfortably let alone 1500.
 
Last edited:
Davedree said:
Lol behave, it's all relevant. Amd rescaling kaveri into carrizo both using the same.process 28shp, is no different to tahiti 28nm rescaling to your denser hawaii architecture using the same 28nm and then concluding it's 100mm smaller than it should be. So don't know why you are claiming a curve ball. It's getting more transistor count into a smaller die.
But you sacrifice clock speed to do it, so you have to have good efficiency in the front end of the gpu.
You will never see a hawaii clock at 1300mhz comfortably let alone 1500.
Click to expand...

Hawaii clocks like Tahiti before it.

Your completely ignoring the power requirements of overclocking, if Hawaii, and Tahiti for that matter requires 400 watts to run at 1500Mhz its never going to do it outside of LN2 because the chip couldn't take the heat on standard cooling, and thats ignoring the stress on the PCB and its components.

That is why you never see any 1500Mhz 290 outside of LN2 benching.

Bring the power requirements down and it will do it, thats why Maxwell does it, nothing to do with the process manufacturing, they are the same for Hawaii and Maxwell.
 
Last edited:
Why you keep comparing hawaii vs tahiti. I'm comparing maxwell vs amd tonga speculation and my own theory of tonga just being scaled up by double. Hawaii isn't hugely different from tahiti, in a way hawaii is wider and in a crude explanation modular like parallel compute.
You said amd don't need to shrink dies, I think you're wrong for thinking that they'll be fine to redraw a denser hawaii into the next gcn on 28nm. Die size is very relevant for yields and business.
They are in my opinion too big considering amd cant demand the money like ibm or nvidia can with big dues. Were the 390's yields so low that amd engineered them so they coild cut them up and sell them as due harvested tonga 1792 2048's. Time will tell

Why can all maxwells clock high, why can a baby maxwell clock high then? No clever power regs on the 750s?
Answer = the architecture, it's all about the front end.
Nvidia have for more efficient shader throughout with less resources. Amd have to use more shaders andbbigger chips to compete.
 
Davedree said:
Why you keep comparing hawaii vs tahiti. I'm comparing maxwell vs amd tonga speculation and my own theory of tonga just being scaled up by double. Hawaii isn't hugely different from tahiti, in a way hawaii is wider and in a crude explanation modular like parallel compute.
You said amd don't need to shrink dies, I think you're wrong for thinking that they'll be fine to redraw a denser hawaii into the next gcn on 28nm. Die size is very relevant for yields and business.
They are in my opinion too big considering amd cant demand the money like ibm or nvidia can with big dues. Were the 390's yields so low that amd engineered them so they coild cut them up and sell them as due harvested tonga 1792 2048's. Time will tell

Why can all maxwells clock high, why can a baby maxwell clock high then? No clever power regs on the 750s?
Answer = the architecture, it's all about the front end.
Nvidia have for more efficient shader throughout with less resources. Amd have to use more shaders andbbigger chips to compete.
Click to expand...

"Why you keep comparing hawaii vs tahiti"

Already explained this.
Because you said denser packing is a problem for AMD in clocking, Hawaii has a lot denser transistor packing than Hawaii and they clock the same.

"You said amd don't need to shrink dies"

I have already explained this too.
Hawaii has about 50% more GPU on a die thats grown by 20% to accommodate it, they have grown significantly without a die shrink, no reason why they couldn't do it again. On current speculative specifications 4096 Shaders a 390X would only be marginally bigger than a Titan-X.

"Why can all maxwells clock high, why can a baby maxwell clock high then? No clever power regs on the 750s?"

Yes there is, the same thing as on big Maxwell.

"Nvidia have for more efficient shader throughout with less resources. Amd have to use more shaders andbbigger chips to compete."

Apples and Oranges again.

AMD's shaders are not the same as Nvidias, AMD have more shaders per die size than Nvidia, Nvidia use a (bigger shader) so to speak.
 
Last edited:
And you still think a die shrink isn't required for AMD?

4096 inefficient shaders vs 3072 efficient nvidia, (getting more done with less)

Time will tell and i'll open a thread to apologise if proven wrong.
2 problems with denser dies, greater chance of defective yields, clocking limited by thermals.

Edit get out clause 390x or whatever amd big die must be run on air,like maxwell.
 
Last edited:
Davedree said:
And you still think a die shrink isn't required for AMD?

4096 inefficient shaders vs 3072 efficient nvidia, (getting more done with less)

Time will tell and i'll open a thread to apologise if proven wrong.
2 problems with denser dies, greater chance of defective yields, clocking limited by thermals.
Click to expand...

Did you read anything above you?

You keep comparing Apples to Oranges.

The GTX 780TI had 2880 Shaders, it was 550mm2 in size, the 290X had 2816 Shaders, its much smaller than the 780TI @ 438mm2, they both have about the same performance and power consumption.

Your problem is you keep looking at the numbers and then stating results that you yourself can't even equate.

In Kepler vs Hawaii, at least, there is no efficiency advantage or process disadvantage for AMD, as a matter of fact AMD get more from less vs Kepler.

Let GCN 1.3 arrive before proclaiming it to be inadequate, that's exactly what was said about a GTX Titan rival, until it arrived and was in some ways better than even i predicted it to be.
 
Last edited:
Stop humbug, we are comparing maxwell to amd's next gpu. Not hawaii vs kepler.
Maxwell efficiency.
Please stop trying to swing it out of perspective or direction.

Edit I agree wait til it comes out then we'll see.
 
Last edited:
HACO said:
Honestly, I'm expecting some dissapointment. A 15-25% increase over 290X seems reasonable, which puts it right below the GTX 980.
Click to expand...

tbh i honesty think the 390x will be between 50-70% than a 290x, most likely 60%... it won't be 100% faster
 
Last edited:
It has to be faster then a 980, no doubt about that, if not it's destined for failure. As lets be honest, a 980 comes out around 30% quicker then a 290x, even less once the resolution is cranked up, If the 390x isn't slapping a 980 about it's going to look silly by the standards set by it's predecessor.

My money is it sitting firmly in the bracket between a 980 and titan x, priced around that of a 980 (£500), amd will play the old 'we have the fastest card' thing again with the 395x2 like they did with the 6990.
 
You must log in or register to reply here.
Back
Top Bottom