GeForce Titan coming end of February

Page 75 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Lonbjerg

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
RussianSensation said:
GTX680's texture fill-rate and compute (shader performance) are being wasted against the 670 which points to either a memory bandwidth or an ROP bottleneck, or both.

1058mhz GTX680 vs. 980mhz GTX670

Shader performance / Gflops = +23%
Texture fill-rate = +23%
vs.
Memory bandwidth = 0%
Pixel fill-rate = +8%

2560x1600 4AA
GTX680 vs. 670 = +10%

2560x1600 8AA
GTX680 vs. 670 = +7%
Source

The increase in the Titan's performance could be greater than the increase in its theoretical functional units against 680 because the 680's functional shaders and texture units are underutilized as a result of 1-2 bottlenecks (ROP and/or memory bandwidth). This is how you may get > 50% over 680 with the Titan in some games at 1600P. <Just my 2 cents>



If you looked at 20+ reviews when HD7970GE launched, it won the majority of them. The conclusion related to noise levels and power consumption was also questionable since AMD stated from the beginning that no retail HD7970GE cards would use the reference blower. It's not AT's fault since AMD didn't send them retail HD7970Ge cards but at the same time the noise levels and power consumption characteristics of those cards had little to do with retail 7970GEs. The launch cards were cooler, quieter and used less power. Additionally, when HD7970GE launched, they were going for $469.99 not $499.99. The review didn't cover any of those points. Not blaming AT directly since other websites also made similar conclusions. It's AMD's fault there for sending reference 925mhz 7970 with flashed BIOS instead of waiting 1 more month to send reviewers actual retail 7970GE cards (which are all after-market).

Other websites did follow-up with proper HD7970 GE reviews testing actual retail versions. HD7970GE used less power than HD7970 925mhz reference card, ran way cooler and quieter.
http://www.techpowerup.com/reviews/VTX3D/Radeon_HD_7970_X-Edition/26.html



If you disagree, it's perfectly fine but it would be appreciated if you did so in a respectful manner. 10,000 units of GK110 doesn't jive with your proposed theory.
Click to expand...


You forgot the link to GK100...
 
B

BallaTheFeared

Diamond Member
Nov 15, 2010
8,115
0
71
You guys done?

GeForce-GTX-Titan-Presentation-7.jpg


Anyone see the bit about the fan?
 
RussianSensation

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
2is said:
Too bad AMD doesn't have an answer for this 1 year old card. But I guess that's a silver lining for them considering they don't have an answer for a 2 year old processor. ;)
Click to expand...

Of course AMD doesn't have an answer. They never intended to since 2006 when they went small die strategy. Did anyone expect AMD to make a 500-550mm2 HD8970 on 28nm? Really? AMD doesn't care about the Titan since HD7000 cards are setting record sales in Europe and January 2013 was the best selling month for that series. When you have limited resources and your CPU/APU/server businesses are drying up, it's a strategic failure to chase < 1% of the market for that firm. Everyone at AMD knows the real battle is 20nm with Volcanic Islands vs. Maxwell. They cannot afford to spend their resources on some marketing exercise. NV can since for NV it's different as GK110 was going to be designed regardless. NV needs large die GPUs for its highly profitable Tesla/Quadro markets. Even if not a single GK110 was to be sold as a GeForce, NV was still going to make one. If anyone at AMD even proposed spending hundreds of millions of dollars to respond, they would have been told no anyway. No one is going to lift a finger to respond to 10K units of a card that was made for workstation markets and also happens to be an awesome gaming chips. For NV it makes sense to continue making these gigantic cards. This is why 8800GTX/280/480/580 were all class-leading cards. This was already discussed back when Fermi launched. NV's and AMD's strategies differ and people seem to not get this just asking for magical cards to show up from AMD.

NV is servicing highly profitable Tesla/Quadro segments with large die chips which allows NV to maintain its key competitive advantages in both GeForce and Professional segments. AMD cannot write-off the millions of dollars of R&D expense that it would cost to create a 500mm2+ chip. Their HPC strategy is only in its infancy. They are trying to get in with OpenCL/GCN but it's been pretty underwhelming effort so far. Anyone who expected AMD to respond with a 500mm2 chip hasn't followed the direction of NV and AMD since 2006. This time it looks like even NV ran into trouble with the large die strategy since in the past they were able to launch hundreds of thousands of 8800GTX/280/480/580 chips but with GK110 they haven't been able to do this in all of 2012 as a GeForce and in 2013 they are barely getting out 10K units. If NV could get GK100/110 out as GeForce in 2012, GTX690 wouldn't even need to exist. They obviously ran into problems and/or it didn't make any financial sense. It probably took them a quarter if not more to build up inventory of 10K GK110 samples to use in GeForce Titan and they started collecting them last year.
 
Last edited:
B

BallaTheFeared

Diamond Member
Nov 15, 2010
8,115
0
71
It sounds like what we already do with MSI Afterburner, so I'm not exactly sure what it means!

Speculation appreciated!
 
Lonbjerg

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
RussianSensation said:
Of course AMD doesn't have an answer. They never intended to since 2006 when they went small die strategy. Did anyone expect AMD to make a 500-550mm2 HD8970 on 28nm? Really? AMD doesn't care about the Titan. Everyone and their mother at AMD knows the real battle is Volcanic Islands vs. Maxwell. If anyone at AMD even proposed spending hundreds of millions of dollars to respond, they would have been told no anyway. No one is going to lift a finger to respond to 10K units of a card that was made for workstation markets. For NV it makes sense to continue making these gigantic cards. This was already discussed back when Fermi launched.

NV is servicing highly profitable Tesla/Tegra segments with large die chips which allows them to maintain its key competitive advantages in both GeForce and Professional segments. AMD cannot write-off the billions of dollars of R&D expense that it would cost to create a 500mm2+ chip. Anyone who expected AMD to respond with a 500mm2 chip hasn't followed the direction of NV and AMD since 2006. This time it looks like even NV ran into trouble with the large die strategy since in the past they were easily able to launch hundreds of thousands of 8800GTX/280/480/580 chips but with GK110 they haven't been able to do this in all of 2012 as a GeForce and in 2013 they are barely getting out 10K units.
Click to expand...

Nice piece of disinformation...but utterly flawed:

http://www.olcf.ornl.gov/titan/

Thats 18,688 GK110 GPU's right there.

You are still going to stick with 10.000 GK110's now?
Alongside "GK100"?

I need to make a list now:

- Uses none-exsistant hardware as "argument"
- Uses false numbers.

Care to add more?
 
Last edited:
moonbogg

moonbogg

Lifer
Jan 8, 2011
10,734
3,454
136
Titan is a novelty and will be a good marketing tool. Tool TOol. That said, I might trade the 670's for one if they are equal or better.
 
2

2is

Diamond Member
Apr 8, 2012
4,281
131
106
RussianSensation said:
Of course AMD doesn't have an answer. They never intended to since 2006 when they went small die strategy. Did anyone expect AMD to make a 500-550mm2 HD8970 on 28nm? Really? AMD doesn't care about the Titan since HD7000 cards are setting record sales in Europe and January 2013 was the best selling month for that series. Everyone at AMD knows the real battle is 20nm with Volcanic Islands vs. Maxwell not wasting financial resources on some marketing exercise. For NV it's different since GK110 was gong to be designed regardless since they needed it for highly profitable Tesla markets. Even if not a single GK110 was to be sold as a GeForce, NV was still going to make one. If anyone at AMD even proposed spending hundreds of millions of dollars to respond, they would have been told no anyway. No one is going to lift a finger to respond to 10K units of a card that was made for workstation markets and also happens to be an awesome gaming card. For NV it makes sense to continue making these gigantic cards. This was already discussed back when Fermi launched. NV's and AMD's strategies differ and people seem to not get this just asking for magical cards to show up from AMD.

NV is servicing highly profitable Tesla/Quadro segments with large die chips which allows them to maintain its key competitive advantages in both GeForce and Professional segments. AMD cannot write-off the millions of dollars of R&D expense that it would cost to create a 500mm2+ chip. Their HPC strategy is only in their infancy. Anyone who expected AMD to respond with a 500mm2 chip hasn't followed the direction of NV and AMD since 2006. This time it looks like even NV ran into trouble with the large die strategy since in the past they were easily able to launch hundreds of thousands of 8800GTX/280/480/580 chips but with GK110 they haven't been able to do this in all of 2012 as a GeForce and in 2013 they are barely getting out 10K units. If NV could get GK100/110 out as GeForce in 2012, GTX690 wouldn't even need to exist. They obviously ran into problems.
Click to expand...

So what is it you're complaining about? nVIdia already has a 680 which competes with a 7970, and in all likelihood, in a few more hours, will have something that completely blows it away.
 
RussianSensation

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
paul878 said:
I thought today was the big day, what happen?
Click to expand...

- NVIDIA GeForce GTX Titan's paper launch (specs and pictures of the card) has been postponed to Feb 19th
- First reviews may not be posted until Thursday the 21st
- The card will be available for purchase no sooner than February 21st, which is a Thursday, probably one of the most common launch dates for NVIDIA
 
Last edited:
Status
Not open for further replies.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom