Maxwell Gen 1

[John P. Myers]

The specs on the 750 and 750Ti are now official, and somewhat hard to believe.

GTX 750Ti:
640 cores
60W TDP
1020MHz base/1085MHz Boost

GTX 750:
512 cores
55W TDP
1020MHz base/1085MHz Boost

These both use the same 28nm GM107 chip.

Performance? Heh. During Nvidia's conference, they showed a slide comparing frames per second while playing Call of Duty: Ghosts.
The HD 4600 IGP scored 10 FPS
A GT 640 (65W version) scored 22 FPS
And the 750Ti scored 51 FPS

This worries me. The performance increase is simply too good in games. When Nvidia went from the 500 to the 600 series, gaming performance increased alot there too, but the 600 series was worse at crunching. Supposedly these new Maxwell chips in the 750 and 750Ti offer 35% more performance per core while using half the power. That's just too much. Something's wrong somewhere.

[DrPop]

When you say, "something's wrong here" do you mean they are trading off Some kind of raw crunching horse power for a certain optimization in games? I'm not sure what that could be, unless it's something like sacrificing details in a scene for higher fps?

[John P. Myers]

Trading off some kind of raw crunching power, yes. Sacrificing game details wouldn't happen though. It's possible Nvidia crippled their FP64 even more, and possibly their INT too. Maybe there aren't even any FP64 units in these cores at all and all INT and FP64 math is converted to FP32 then reconverted to INT or FP64 for output (essentially an emulator). In games you wouldn't notice this at all since they mainly focus only on FP32. In crunching though, it would be horrid.

When someone makes the correct slot configuration of the 750, i'll get one and test it out to see. I suppose there's some slim possibility it really is faster in all areas, i just find it hard to believe at the moment. There's been no die-shrink here. It's the same process used since the 600 series. Maxwell Gen 2 will have the die-shrink to 20nm and i'd be much more likely to believe these numbers then, but without the die-shrink claiming to have 35% more performance per core at half the TDP is a huge leap for a single generation. After the 20nm Maxwells start showing up, how much of an additional performance increase over Gen 1 are we looking at? lol it's just too good to be true.

[John P. Myers]

Here's a comparison - GK107 (Kepler) vs. GM107 (Maxwell)

The GM107 die is 26% larger, has 256 more cores, has 570 million more transistors, 10% more bandwidth, is rated at 494 *more* GFLOPS FP32 (a whopping 61% increase), but uses 6.25% less power to do all this. Oh and 400MHz faster VRAM too (an 8% increase).

So Nvidia is telling me i can crunch 61% faster using slightly less electricity in just 1 generation? I call BS :p

BTW these GPUs are launching in 3 days.

[DrPop]

...
So Nvidia is telling me i can crunch 61% faster using slightly less electricity in just 1 generation? I call BS :p

BTW these GPUs are launching in 3 days.

Yeah, maybe it is too good to be true, but man that would be nice. :)

[STE\/E]

I always thought the amount of core's was the most important for crunching, they don't appear to have that many cores for BOINC Crunching ...

GTX 750Ti:
640 cores
60W TDP
1020MHz base/1085MHz Boost

GTX 750:
512 cores
55W TDP
1020MHz base/1085MHz Boost

[pinhodecarlos]

I think a better energy efficiency card as in a normal energy system is the way to go. Maybe it is not a BS but true. Nvidia is going in the right direction when energy consumption is concerned. For the last 3 years I've been an energy analyst and that's the way people in the area do, they try for the same about of electricity to produce more. Take my hat off to Nvidia if it comes true.

[John P. Myers]

I always thought the amount of core's was the most important for crunching, they don't appear to have that many cores for BOINC Crunching ...

No they don't have many cores as this is a lower-midrange GPU. There are a few reasons i'm really interested in this though. It's the first chip based on the Maxwell architecture and since it's not a high-end GPU, the performance of the 750Ti will actually dictate how the rest of the Maxwell GPUs (the GTX 800 series) perform. Efficiency always increases as you move to the larger more expensive GPUs and with the 750Ti's GFLOPS/W rating starting out higher than even a GTX 780Ti, it follows that the rest of the 800 series would completely obliterate anything else ever made.

Another thing about the GPUs is you could buy 8 of the GTX 750Ti's for about the price of a Titan but compute 2.32 times faster. Another reason i like looking at these lower-end cards is the lower power draw allows manufacturers to make them only 1 slot wide, and i like finding GPUs i can stick in empty slots just as fillers. Currently this is the fastest, most power efficient GPU that is only 1 slot wide: http://www.newegg.com/Product/Product.aspx?Item=N82E16814125468 I was first to review it :D If the 750 and 750Ti are as good as Nvidia says they are, they will become my new #1 choice.

[Al]

So, Low Profile, which means it will fit in my servers. Looks like a decent upgrade to the 430s that are in there now.

[John P. Myers]

So, Low Profile, which means it will fit in my servers. Looks like a decent upgrade to the 430s that are in there now.

Yes, much better, but don't get one just yet. With the GTX 750 rated at only 55W, it should also be possible to make one low profile and *should* outperform the 640 rev2 by better than 25%.

[Al]

Well, you just stopped me from buying 2. Please keep us update...as if you wouldn't. :)

[Fire$torm]

No they don't have many cores as this is a lower-midrange GPU. There are a few reasons i'm really interested in this though. It's the first chip based on the Maxwell architecture and since it's not a high-end GPU, the performance of the 750Ti will actually dictate how the rest of the Maxwell GPUs (the GTX 800 series) perform. Efficiency always increases as you move to the larger more expensive GPUs and with the 750Ti's GFLOPS/W rating starting out higher than even a GTX 780Ti, it follows that the rest of the 800 series would completely obliterate anything else ever made.

Another thing about the GPUs is you could buy 8 of the GTX 750Ti's for about the price of a Titan but compute 2.32 times faster. Another reason i like looking at these lower-end cards is the lower power draw allows manufacturers to make them only 1 slot wide, and i like finding GPUs i can stick in empty slots just as fillers. Currently this is the fastest, most power efficient GPU that is only 1 slot wide: http://www.newegg.com/Product/Product.aspx?Item=N82E16814125468 I was first to review it :D If the 750 and 750Ti are as good as Nvidia says they are, they will become my new #1 choice.

But, can you water cool it...? =))

[Duke of Buckingham]

Can we cool it with beer?b-(

[John P. Myers]

However, double-precision math is further pared back to 1/32 the rate of FP32; that was 1/24 in the mainstream Kepler-based GPUs.
$%^$%*! Nvidia

[conf]

Not very good for crunching. In Europe the 750ti costs the same as the "old" 660GTX,
but the second one has 150% power related to the new maxwell.
For energy purposes the new one is better but both have low power consumption.

plus a question : if I have two ATIs running how do I add a Nvidia in third PCIe port for different projects like Einstein or GPUGrid,
tried this often before but it never worked as I had expected. Used a cc_config for these trials.

[John P. Myers]

If you're using windows vista you can't. Must be xp, 7 or 8. In linux I'm not sure. That's someone else's specialty.

Edit: fp64 has always been horrible from nvidia. People use them for the fp32.

Sent from my Galaxy S4 using Tapatalk Pro

[conf]

All my boxes run with Windows 7 or 8.
I had the problem that the strong GPUs ran the slow projects ,as Einstein is, after a few hours or days,
The small one made less problems and I was able to fix it to one project.
I could use a slow Ati without any problems but this makes no sense to me.
The third slot should only be used for this half Gpu projects like seti,einstein or wcg.

[John P. Myers]

plus a question : if I have two ATIs running how do I add a Nvidia in third PCIe port for different projects like Einstein or GPUGrid,
tried this often before but it never worked as I had expected. Used a cc_config for these trials.

No cc_config needed. Just install the Nvidia GPU, install drivers, then select the project you want to run and tell it to only use the Nvidia GPU, then you won't get AMD/ATI WUs from that project. The projects you want to run on AMD/ATI, don't allow them to use Nvidia.



After 3+ hours of digging and digging for real numbers and more info, i found 2 things important to note :)

First is that manufacturers (most likely Zotac and Gigabyte) have stated they will be making true single-slot versions of not only the GTX 750, but of the 750Ti as well. Game on! You can assume from this there will also be a LP version of at least the GTX 750 for Al :p

Second is it *does* appear compute has increased a good bit. I saw FAH Bench results that we proming (Folding at Home's benchmark). I also saw benchmarks in regards to alt-coin mining. Not what we're doing, but math is math. The results were very promising in both cases. When the tru single-slot versions come out, i will buy one that day and get some testing done.

Also thirdly, Nvidia seems to have underclocked these cards big time. People are reporting stable results at 1300MHz with a tiny 31mV bump. EVGA has a version of the 750 with a 1294MHz Boost out of the box (stock is 1085MHz). That's pretty significant.

[Fire$torm]

For crunching, an ATI/AMD GPU MUST occupy the first PCIe slot closest to the CPU. Then the 2nd & 3rd slots can be home to CUDA cards.

[zombie67]

So, where are we on this? Any of these new cards single slot?

[John P. Myers]

So, where are we on this? Any of these new cards single slot?

Not yet :/ Believe me, i'm checking daily. I did have a chat with EVGA a couple days ago and asked about it. I was told they won't be making a single slot, which isn't surprising because they never do anyway on the lower cards. Pretty much what i expected. I did find out that the Titan Black will have a liquid version coming soon with the waterblock made by EK.

When the single-slots happen, 99% chance it'll be by Zotac. Maybe 70% chance from Gigabyte.

Edit: The reason it's taking so long is Nvidia was nice enough to redesign the screw hole spacings so any available cooler no longer fits. They're now 44 x 44mm when previously they've been 53.2 x 53.2mm or 58.4 x 58.4mm. So now all the GPU manufacturers have to design new heatsinks from scratch. Nvidia screws us over once again :/

[zombie67]

What about AMD? How good is their best single slot GPU?

[John P. Myers]

What about AMD? How good is their best single slot GPU?

The HD 7750 is AMd's best which is about equal to a GT 640 Rev2.

[conf]

Sapphire Radeon R7 250 ??
http://geizhals.at/sapphire-radeon-r7-250-11215-06-20g-a1070502.html
Seems to be equal but is a new card from 2014.

[FourOh]

What about AMD? How good is their best single slot GPU?


The HD 7750 is AMd's best which is about equal to a GT 640 Rev2.


Sapphire Radeon R7 250 ??
http://geizhals.at/sapphire-radeon-r7-250-11215-06-20g-a1070502.html
Seems to be equal but is a new card from 2014.

Yes, best single slot from the Radeon series will be 7750 or R7 250 (or possibly 6850), but the FirePro series has some way more powerful single-slot cards. Best single-precision is FirePro W7000 with 2432 GFLOPS, best SP/DP card is V7900 with 1860/464 GFLOPS.

[Fire$torm]

XFX FX-775A-ZNP4 Radeon HD 7750 Core Edition

Amazon: $100.74 w/Free Shipping (http://www.amazon.com/XFX-DisplayPort-PCI-Express-FX775AZNP4-FX-775A-ZNP4/dp/B007Z3T5JC)

[conf]

The FirePros are too expensive anyway. Maybe one can find a used one on Ebay.

Edit. one v7900 found at Ebay for 399,- Euro, not so bad as I thought it would be.

[Al]

So, are any of these going to be available in, 1) low profile and 2) no need for and additional power connector? I'm looking for the best card to put in my servers to replace my aging 430s.

[Fire$torm]

So, are any of these going to be available in, 1) low profile and 2) no need for and additional power connector? I'm looking for the best card to put in my servers to replace my aging 430s.

The XFX unit I linked to doesn't require any PCIe power cables. Found a nice overview vid of that card (Here (http://www.youtube.com/watch?feature=player_embedded&v=yCYPos4C5EY))

Edit: It looks like there is currently a premium attached to the low-profile versions of the 7750. VisionTek's is $30~$40 more then the XFX single slot.

[Al]

I assume the low profile kit he mentioned is a low profile kit for that card? If so, that would work. I'll look around and see if I can find anything to give me an idea on the performance increase I might expect over the 430.

[FourOh]

I assume the low profile kit he mentioned is a low profile kit for that card? If so, that would work. I'll look around and see if I can find anything to give me an idea on the performance increase I might expect over the 430.

I run a low-profile Sapphire HD 7750 in a Dell box - it was great on the old Poem OpenCL app; now have it on Collatz where it puts out about 200k/day. Output on DiRT is comparable.

[Fire$torm]

I assume the low profile kit he mentioned is a low profile kit for that card? If so, that would work. I'll look around and see if I can find anything to give me an idea on the performance increase I might expect over the 430.
Me thinks that was just a sales brochure. That XFX card is too tall for low-profile

[conf]

Besides the one slot design there is a new powerful but cheap Ati card for the third PciE slot:
Sapphire Radeon R7 260X OC, 1GB GDDR5 (115W , 2061GFLOPS (Single), 129GFLOPS (Double)It costs only 135 $ in Europe but takes two slots.

[John P. Myers]

So, are any of these going to be available in, 1) low profile and 2) no need for and additional power connector? I'm looking for the best card to put in my servers to replace my aging 430s.

If you're asking about the GTX 750/750Ti, they are both available without additional power connectors. Looks like it will be a bit yet before any true single slots and low profiles are made though, since Nvidia changed the heatsink screw hole spacings causing heatsinks to be redesigned from scratch.

[John P. Myers]

Galaxy made a low profile GTX 750 Ti :D Only problem is, i can't find it anywhere. Single bracket but heatsink/fan is 2 slots wide. Still it may work in some slim/server cases. http://www.galaxytech.com/__EN_GB__/Product2/ProductDetail?proID=519&isStop=0&isPack=False&isPow=False

Edit: nvm found galaxy's online sales. Here it is: http://store.galaxytechus.com/GALAXY-GeForce-GTX-750-Ti-GC-Slim-2-GB-GDDR5-PCI-Express-30-DVIHDMIVGA-Graphics-Card-75IGH8HX9KXZbrbrFree-low-profile-brackets-included-limited-time-offer_p_90.html

And the GTX 750 also low profile: http://store.galaxytechus.com/GALAXY-GeForce-GTX-750-GC-Slim-1-GB-GDDR5-PCI-Express-30-DVIHDMIVGA-Graphics-Card-75NGH8HX9KXZbrbrFree-low-profile-brackets-included-limited-time-offer_p_89.html

[Al]

Cool! That might actually work for me, LP and no 6-pin necessary. I'll need to do some measuring first.

[zombie67]

Interesting thread here: http://einstein.phys.uwm.edu/forum_thread.php?id=10613

Very high load with a single task on einstein. Also, they are coming to the conclusion that the Ti is not worth the extra cost or electricity.

[Al]

Cool! That might actually work for me, LP and no 6-pin necessary. I'll need to do some measuring first.

No good, the cooler is too tall. Needs to be a true single slot, low profile and no additional power connector and an upgrade to a GT430. I could possibly mod the server case and use pcie extenders, but that is not my first choice.

[FourOh]

Interesting thread here: http://einstein.phys.uwm.edu/forum_thread.php?id=10613

Very high load with a single task on einstein. Also, they are coming to the conclusion that the Ti is not worth the extra cost or electricity.

It is an interesting thread and for Einstein it appears that with current drivers the additional cost of the Ti version isn't worth the added performance. However, based on a small sample, the 750 Ti outperforms the 750 by roughly 15% on GPUGrid, translating to and additional 25-30,000 credits per day. To me, particularly considering the additional 1Gb GDDR5 on the Ti version, that's worth the additional $40 outlay.

[John P. Myers]

Yeah i don't trust Einstein benchmarks at all. For me, my GTX 580 is 3x faster than my Titan....but only on Einstein.

[DrPop]

...To me, particularly considering the additional 1Gb GDDR5 on the Ti version, that's worth the additional $40 outlay.

I think that may be the ticket for future proofing in a way. I remember when the projects required 512MB RAM or they wouldn't run right. At some point it's not inconceivable that projects would require X amount of RAM or you can't get a WU - and the Ti version might win out in the long term.
Of course if you rotate your hardware every year or two then that's probably not a big concern yet. ;)

[FourOh]

Screw it. I just ordered one: http://www.evga.com/Products/Product.aspx?pn=02G-P4-3753-KR
It will replace the HD 6570 in my HTPC - about double the GFLOPs and 4-5x the performance on most BOINC apps for the same 60 Watts!

[Z-TAC]

Screw it. I just ordered one: http://www.evga.com/Products/Product.aspx?pn=02G-P4-3753-KR
It will replace the HD 6570 in my HTPC - about double the GFLOPs and 4-5x the performance on most BOINC apps for the same 60 Watts!

Way to step up to the plate! :)

[Mumps]

So, for $124, you're saying this Zotac (http://www.newegg.com/Product/Product.aspx?Item=N82E16814500326) would be a safe bet for a basic cruncher for things with wimpy power supplies? As long as it has even PCIe 1.1?

[John P. Myers]

So, for $124, you're saying this Zotac (http://www.newegg.com/Product/Product.aspx?Item=N82E16814500326) would be a safe bet for a basic cruncher for things with wimpy power supplies? As long as it has even PCIe 1.1?

Correct. Or this Gigabyte one: http://www.newegg.com/Product/Product.aspx?Item=N82E16814125503 Same price (almost) but just a little faster.

[Al]

So it would appear, from a power required standpoint, I would be better off to sell my 3-570s and my 560ti 448 core and replace them with 750s or 750ti's. The power draw decrease is dramatic, which would be nice. The specs I've seen show more cuda cores on the 750ti, but that doesn't necessarily equate to better crunching ability. Do we know how the 750ti crunches by comparison to the 570? Just looking at PrimeGrid's Fastest GPUs seems to indicate the the 750ti is considerably slower than the 570. That isn't necessarily a deal killer, just curious.

Edit: Comparison from PrimeGrid. I have no idea if the 750ti is oced or not, my 570 is to 850...but the 750ti is still only 60 watts.

Detected GPU 0: GeForce GTX 750 Ti
Detected compute capability: 5.0
Detected 5 multiprocessors.

Thread 0 completed
Sieve complete: 377188221000000000 <= p < 377188230000000000
count=222379442,sum=0xaa8099dfa067f758
Elapsed time: 1970.24 sec. (2.97 init + 1967.27 sieve) at 4574974 p/sec.
Processor time: 151.98 sec. (2.78 init + 149.20 sieve) at 60323249 p/sec.
Average processor utilization: 0.94 (init), 0.08 (sieve)
19:06:50 (1164): called boinc_finish
================================================
Detected GPU 1: GeForce GTX 570
Detected compute capability: 2.0
Detected 15 multiprocessors.

Thread 0 completed
Sieve complete: 379257411000000000 <= p < 379257420000000000
count=222360166,sum=0xe318f766e0117a18
Elapsed time: 815.44 sec. (2.50 init + 812.94 sieve) at 11071146 p/sec.
Processor time: 113.16 sec. (2.50 init + 110.67 sieve) at 81326693 p/sec.
Average processor utilization: 1.00 (init), 0.14 (sieve)
09:25:33 (1120): called boinc_finish

[Fire$torm]

So it would appear, from a power required standpoint, I would be better off to sell my 3-570s and my 560ti 448 core and replace them with 750s or 750ti's. The power draw decrease is dramatic, .......

Well if my math is correct, based on those numbers you posted the 570 potentially uses 3.65 times more power to do 2.42 times the work. But that is only one project. My results most likely may not apply to say, a project like Collatz.

#5 needs more input...

[Al]

Well, here's some other projects for comparison. (http://www.overclock.net/t/1472363/gtx-750-ti-boinc-performance) Someone else do the math. Apparently it sizes up to a 480, but at only 60 watts. I guess it boils down to whether you can accept less cpd to save on the electricity. Of course if you put 3 or 4 single slots (if they become available) in a system you might not really be saving much.

[Fire$torm]

Well, here's some other projects for comparison. (http://www.overclock.net/t/1472363/gtx-750-ti-boinc-performance) Someone else do the math. Apparently it sizes up to a 480, but at only 60 watts. I guess it boils down to whether you can accept less cpd to save on the electricity. Of course if you put 3 or 4 single slots (if they become available) in a system you might not really be saving much.

Sure you would. x3 750 = 180W max. 570 = 219W max. That's the equivalent of a 40W light bulb. Plus, those 3 750s will out perform that single 570. Another side benefit is redundancy. If a 750 fails, you still have the other two crunching away. Not so with a single 570.

The only drawback is the MB. To handle 3 of any kind of GPU requires a MB with dual 4-pin or 6-pin 12V Aux. connectors. Just too much current draw for a single Aux.

[John P. Myers]

If the 750Ti in the 570 comparison was a base model, then you could very easily require only 2 of them to equal the 570 for a whole let less wattage. Some of the OC models are 20%+ above base clocks out of the box, and can still be OC'd further, though these usually have a 6-pin plug added just in case. You do not have to plug anything into it if you don't want to. The heatsinks on the 750/750Ti are overkill so it will still run cool even if you jacked it up to 1300MHz.

so basically the math says, if you have the slots available, you can replace your 570 with 2 or even 3 750Ti's, crunch faster, produce less heat and save money on your bill all at the same time. Oh and it'd be quieter too.

[FourOh]

From the looks of this GTX 750 Ti on Collatz, expected results should be around 300k/day:
http://boinc.thesonntags.com/collatz/results.php?hostid=133526

That's 50% more production than the HD 7750 (currently the fastest AMD card that doesn't require auxiliary power) with 20% less power draw.

I'm hoping to get my 750 Ti up and running tonight or tomorrow morning and start putting it to the test!

[DrPop]

Wow! That's pretty good. :) For those of us "utility bill conscious" crunchers, please keep us posted on the results.

[John P. Myers]

I did find out that the Titan Black will have a liquid version coming soon with the waterblock made by EK.
I guess the EVGA rep was mistaken. The Titan Black Hydro Coppers have been released and have Swiftech waterblocks.

I've been trying to find a way to liquid cool a GTX 750Ti to make it a true single slot with no luck, but maybe there are some northbridge blocks that will work on it :p It's too tiny!

[John P. Myers]

I've been trying to find a way to liquid cool a GTX 750Ti to make it a true single slot with no luck, but maybe there are some northbridge blocks that will work on it :p It's too tiny!

We have a winner!! (soon)


EK Water Blocks is also working on a new value oriented water block for the NVIDIA® Maxwell GM107 GPU based GeForce® GTX 750- and GTX 750 Ti series graphics cards. EK-FC750 GTX will again feature single-slot design to free adjacent PCI(e) slot and allow stacking of these graphics cards. Due to variety of NVIDIA® GeForce® GTX 750 (Ti) circuit board designs and no prescribed standard this water block will directly cool GPU and RAM only in order to maximize hardware compatibility.

[FourOh]

I've been running 750 Ti on Collatz for a couple days; results are 300k/day as expected. At 60W that's 5000 credits/day/watt - very strong, just edging out the HD 7790 which produces about 4700 credits/day/watt.

[myshortpencil]

I've been running 750 Ti on Collatz for a couple days; results are 300k/day as expected. At 60W that's 5000 credits/day/watt - very strong, just edging out the HD 7790 which produces about 4700 credits/day/watt.

Stupendous! Have you seen How to Increase the GeForce GTX 750 Ti Power Target Limit (http://cryptomining-blog.com/1014-how-to-increase-the-geforce-gtx-750-ti-power-target-limit/)?

[FourOh]

Stupendous! Have you seen How to Increase the GeForce GTX 750 Ti Power Target Limit (http://cryptomining-blog.com/1014-how-to-increase-the-geforce-gtx-750-ti-power-target-limit/)?

TWEEEEEEAK

I'll have to give this a try soon, before the weather starts to warm up (any week now). My HTPC is in the back of a closet with very little ventilation, so I don't want to push it.

[John P. Myers]

And poof. Just like that the Denver CPUs that were supposed to appear in the GTX 800 series GPUs have vanished with no explanation from Nvidia. In fact, they're acting as if they never existed in the first place. No more ARM chips on the GPUs to help speed things up afterall.

[Fire$torm]

nVidia must have a really big magic wand...

[John P. Myers]

Bad news. Looks like TSMC is failing constantly (still) and won't be able to make 20nm GPU chips for AMD or Nvidia for quite awhile yet. Without it, Nvidia's Maxwell's won't be nearly as tempting. Same for AMD. TSMC has been trying for quite awhile now and they just keep screwing it all up :(

[Fire$torm]

Bad news. Looks like TSMC is failing constantly (still) and won't be able to make 20nm GPU chips for AMD or Nvidia for quite awhile yet. Without it, Nvidia's Maxwell's won't be nearly as tempting. Same for AMD. TSMC has been trying for quite awhile now and they just keep screwing it all up :(

What is their issue? Are the yields too low or are they unable to fab the wafers at all?

[finalfugue]

Bad news. Looks like TSMC is failing constantly (still) and won't be able to make 20nm GPU chips for AMD or Nvidia for quite awhile yet. Without it, Nvidia's Maxwell's won't be nearly as tempting. Same for AMD. TSMC has been trying for quite awhile now and they just keep screwing it all up :(

Sooo, on the plus side my 750 Ti's will still look pretty sweet for some time to come?

Another one (larger, more fans, FTW edition) on a special approximately the same price as the last "short" superclocked one at NewEgg:

http://www.newegg.com/Product/Product.aspx?Item=N82E16814487028&nm_mc=EMC-IGNEFL042214&cm_mmc=EMC-IGNEFL042214-_-EMC-042214-Index-_-DesktopGraphicsVideoCards-_-14487028-L0D

[John P. Myers]

What is their issue? Are the yields too low or are they unable to fab the wafers at all?

Yep yields. They've been working on it for like 2 years now and still suck. Meanwhile Intel, Samsung and Global Foundaries are making stuff at 14nm right this second, laughing their asses off at TSMC. TSMC has always seemed to me to be the dumb kid in the group, but this has gone way past that into the realm of incompetence. Word has it Nvidia had to rework some of their Maxwell chip designs to stay at 28nm. They're not happy in the least. If they can't get someone who has a clue what they're doing, TSMC may end up having to shut their doors over this eventually. And even if they do figure it out, i shudder to think what will happen when they try the next smaller die shrink after 20nm.

Pisses me off. I hate Nvidia, but i've been looking forward to some good power efficient Maxwell GPUs :mad:

[zombie67]

I hope TSMC resolves this and eventually catches up to their competition. Less suppliers is not good for us.

[Fire$torm]

Sounds like TSMC needs to stop being cheap and hire some intelligent talent at ridiculously high wages. Long term, they would lose a LOT LESS money. I hate bean counters. (Quasi apologies to any bean counters on the forums).

[finalfugue]

Looks like there's quite a bit of turmoil in the foundries world ATM:

http://techreport.com/news/26336/globalfoundries-licenses-samsung-process-tech-grants-amd-access-to-finfets

Would love to be a fly on the wall in a few executive offices right now. Wonder who is going to move where, when? I'd be more disappointed about the new Maxwell delay, but considering I'm pretty broke now, part of me is glad my wallet has more time to recover :P.

[zombie67]

Wasn't there some noise about Apple building their own chips? Did that mean their own factory? Or just taking their own design from Samsung to some place like TSMC? Apple's volumes are a pretty large share of the market. And another manufacturer would be a good thing.

[DrPop]

I thought I read somewhere about Intel being willing to start opening up their fabs for bidders . . . I have nothing to back that up but a vague memory of some online article, though. Wonder if Intel will sell some fab time to NVidia? Highly doubt they would sell AMD any, even at top dollar though! :P

[John P. Myers]

Wasn't there some noise about Apple building their own chips? Did that mean their own factory? Or just taking their own design from Samsung to some place like TSMC? Apple's volumes are a pretty large share of the market. And another manufacturer would be a good thing.

Yeah Apple is expected to have TSMC make their 20nm processors, then switch to Global Foundries for their 14nm processors so they can pat themselves on the back for not using Samsung processors even though GF is making the processors with tech they licensed from Samsung. Good job there, Apple :rollseyes: I also heard Apple was going to start making their own chips and that they've "bought into" a fab. Whether that's for mass production or so they can more easily create/test designs i'm not sure yet, though going from nothing to making all of their own chips, especially at Apple's volume, seems like an incredibly daunting task. Hopefully for them the results would be better than their attempt at maps :/ However, i've also heard Apple wants to make their own GPUs. Maybe that's their intention? Maybe both? I guess we'll see.

Something i've seen no one mention though is with Apple using TSMC for 20nm instead of Samsung, it's now just 1 more (major) thing of Apple's not made in America, since Samsung was making Apple's processors in Austin, TX. So much for Apple's "Made in the USA" PR media storm they started last year :p

Also even though TSMC is sucking it hard trying to make 20nm GPU wafers, CPUs are easier to make so i don't really see a terribly high error rate there, though it won't take much of one to really cripple their ability to produce enough volume.

TSMC was supposed to've had 20nm ready in time for the GTX 750/750 Ti and for AMD's R9 290/290X. They didn't, but these companies had to release something. TSMC's 20nm abilities are so awful that even with the chips that do work, they're less efficient than the 28nm chips.

[Fire$torm]

Looks like there's quite a bit of turmoil in the foundries world ATM:

http://techreport.com/news/26336/globalfoundries-licenses-samsung-process-tech-grants-amd-access-to-finfets

Would love to be a fly on the wall in a few executive offices right now. Wonder who is going to move where, when? I'd be more disappointed about the new Maxwell delay, but considering I'm pretty broke now, part of me is glad my wallet has more time to recover :P.

Wow, there had to be a ton of closed door wrangling goings on to make all that come together....

[finalfugue]

There are more secret deals and shifting alliances than in "Game of Thrones" right now.

House AMD- what are your words? "Keep Hell Warm. (http://www.techpowerup.com/reviews/AMD/R9_295_X2/22.html)"

[MindCrime]

I thought I read somewhere about Intel being willing to start opening up their fabs for bidders . . . I have nothing to back that up but a vague memory of some online article, though. Wonder if Intel will sell some fab time to NVidia? Highly doubt they would sell AMD any, even at top dollar though! :P

Intel is a rare breed; to design and build their own CPUS. Barely anyone makes their own CPU or GPU, especially mobile. But Intel does, they were always that way and will stay that way. Fabs have 1000s of people passing through them daily, vendors, contractors, and actual Fab (intel, tsmc, gloFo) employees. The processes are coveted, lithography can be stolen, intel is copy exact super anal, no way they'll let someone else make their CPUs and I doubt they'll share fab time, seems like they're always building a new one.