Message boards : Graphics cards (GPUs) : GeForce GTX Titan launching the 18th
Author | Message |
---|---|
There's an interesting artical here | |
ID: 28588 | Rating: 0 | rate: / Reply Quote | |
Click instead | |
ID: 28589 | Rating: 0 | rate: / Reply Quote | |
Heres some pictures of it here | |
ID: 28590 | Rating: 0 | rate: / Reply Quote | |
Looks nice, love the window. I think I'd put it on a hinge or a slider like a sunroof to make dust blow outs easier. | |
ID: 28592 | Rating: 0 | rate: / Reply Quote | |
Now their saying delayed untill the 19th, heres a whole new batch of pictures. | |
ID: 28593 | Rating: 0 | rate: / Reply Quote | |
According to its specifications, it will be ~39.5% faster than a standard GTX 680. Probably it will be more power effective. | |
ID: 28598 | Rating: 0 | rate: / Reply Quote | |
At that price point 10k might actually be enough :D | |
ID: 28601 | Rating: 0 | rate: / Reply Quote | |
Good looks and limited edition aren't worth that much to me, maybe in a car/truck but not a video card. | |
ID: 28607 | Rating: 0 | rate: / Reply Quote | |
mostly agree, it's a component... specs matter the most. however sometimes it's worth it for peripherals (keyboard, mouse, things you actually SEE and TOUCH on a regular basis :P) | |
ID: 28608 | Rating: 0 | rate: / Reply Quote | |
NVIDIA's GeForce GTX Titan, Part 1: Titan For Gaming, Titan For Compute | |
ID: 28616 | Rating: 0 | rate: / Reply Quote | |
I guess disabling FP64 (running at 1/24) would be better for here, but that depends on the apps and WU's. GPU-Grid uses exactly 0 FP64, so disabling won't hurt. And not disabling Turbo by leaving FP64 at 1/24 will benefit GPU-Grid, depending on what other manual OC one might have in place instead. Crippling FP64 to get Turbo s*cks hard for people who want to run a DP backup project, though. Generally I think nVidia is being stupid forcing this decision on us. Sure, the chip will consume more power if fully utilized.. but they've got power consumption monitoring and temperature control in place, as well as fine-grained Turbo modew with voltage adjustement. I can't see any harm in leaving Turbo and FP64 at 1/3 active and just throttling down to base clock & voltage if really neccessary. This sure beats starting at base clock & voltage irregardless of actual workload. What if anything would Dynamic Parallelism, Hyper-Q and more Registers/Thread bring to this project? More registers/thread avoid slowing down your more complex code. However, if the code is written to perform well on regular Keplers (and older), then you should automatically stay out of the region where this would matter. The other 2 features have to be explicitely programmed for, so are not going to be used here unless all GPUs get them. MrS ____________ Scanning for our furry friends since Jan 2002 | |
ID: 28619 | Rating: 0 | rate: / Reply Quote | |
Unless the researchers can find something special hidden within this card I'm writing it off as a niche GPU; not for general users, gamers and 'normal' GPU crunchers. It's better suited to other things, mainly due to the price. While the Titan is essentially a Tesla for ~£830 (which will be excellent for some researchers) I cannot see GPUGrid needing 6GB GDDR5, and I can't see the uptake of this card being high enough for GPUGrid to develop specifically for it, especially if there will only be 10,000 of them - If GPUGrid started to go down that route, the project would need about 7 different apps at any one time (probably based on CC). We really need to see what the performance actually is for GPUGrid (and other projects) before drawing any solid conclusions. That said, it's clear that the price is exclusively high. | |
ID: 28622 | Rating: 0 | rate: / Reply Quote | |
The biggest advantage I see for the GTX Titan is the exhaust system. It would better facilitate 3 or 4 GPU's in the one case: What makes Titan's exhaust system better? Why is 3 GTX690s impractical? Impractical from a cooling/exhaust perspective? If so why? (I plan on ordering a mobo and single GTX690 at month's end and plan on eventually adding 3 more GTX690 to the same mobo. If you think it's a bad idea from an exhaust/cooling perspective or any other perspective I'd appreciate hearing about it now before I make the initial purchase.) ____________ BOINC <<--- credit whores, pedants, alien hunters | |
ID: 28623 | Rating: 0 | rate: / Reply Quote | |
gtx 690 exhausts inside the pc case (at least half anyways). most other designs (including 680/titan) exhaust outside the pc case. This is only referring to the reference design though; many 3rd party manufacturers change the cooling systems to something completely different. | |
ID: 28632 | Rating: 0 | rate: / Reply Quote | |
Sure, this card is a niche product. It's for people who don't want to or can't afford a Tesla and make a living from crunching things on it and need either massive FP64 or can't parallelize the task well. Or just want an über-gaming rig and don't care about the money :p | |
ID: 28642 | Rating: 0 | rate: / Reply Quote | |
NVIDIA's GeForce GTX Titan, Part 1: Titan For Gaming, Titan For Compute NVIDIA’s GeForce GTX Titan Review, Part 2: Titan's Performance Unveiled - compute performance benchmarks. Well written by a CS PhD student with a specialization in parallel computing and GPUs. Unrivaled compute performance here in both DP and SP. IMHO, nVidia got wise with this card. I bet it flys of the shelves to people who want this kind of compute performance at a fraction of the cost of their Teslas. If only my wallet were large enough... ____________ | |
ID: 28662 | Rating: 0 | rate: / Reply Quote | |
Unless the researchers can find something special hidden within this card I'm writing it off as a niche GPU; not for general users, gamers and 'normal' GPU crunchers. Agreed. ____________ | |
ID: 28663 | Rating: 0 | rate: / Reply Quote | |
http://www.anandtech.com/show/6774/nvidias-geforce-gtx-titan-part-2-titans-performance-unveiled | |
ID: 28664 | Rating: 0 | rate: / Reply Quote | |
What will be better...2xGTX690 (4 GPU total) or 2xGTX Titan? | |
ID: 28673 | Rating: 0 | rate: / Reply Quote | |
I might end up saving up and grabbing one for gaming. I have a single GTX 670 right now and it's not quite cutting it. Some of the games I play don't really scale very well with multiple GPUs, and upgrading to a 680 isn't worth the time or trouble. If I was patient I'd just wait for the new chips to come out in 2014 but that's not the case. Getting insane compute performance out of it is just an added bonus! | |
ID: 28676 | Rating: 0 | rate: / Reply Quote | |
What will be better...2xGTX690 (4 GPU total) or 2xGTX Titan? Two Titan's would be more expensive and do less work (in theory). The benefit of a Titans is the reference exhaust cooling system over the radial system used by the GTX690; which blows air out of the back of the case, but also into the case. A single GTX690 is fine, but multiple GTX690's would require you to have a very large side panel input fan and exhaust both from the rear and front of the ATX case. This is probably doable for two GTX690 cards, but for 3 or 4 cards it's more difficult. Also if you intend to use Linux you could in theory have 8 GPU's (four GTX690 cards) but in Windows I don't think you can. Basically GPU's don't scale well. Adding more GPU's requires greater skills, and isn't something for the amateur cruncher. If anyone is seriously considering 3 or more GTX690's you should primarily be thinking about cooling. Something like this case (with 4 side panel fans) could be used to blow cool air onto the GPU's, and then it could be drawn out from the rear and front. I'm sure this sort of case could handle two GTX690's but I would want to know the temperatures before adding a third, and would only add them one by one. I also like the one very large side panel fan on this case, but not the drive bays. This open frame case looks very interesting, if you don't mind dusting. ____________ FAQ's HOW TO: - Opt out of Beta Tests - Ask for Help | |
ID: 28677 | Rating: 0 | rate: / Reply Quote | |
What will be better...2xGTX690 (4 GPU total) or 2xGTX Titan? From the performance per price point of view the GTX690 is better than a Titan, as BOINC loads scale pretty much perfectly with GPU count (except POEM). However, if you're running into limits with the amount of GPUs the Titans might be a better option. Not for 4 vs. 2 chips, though. And one more point for the Titan: it will be able to crunch GPU-Grid tasks within the bonus time for longer. If the power consumption still allows to run the card when it's in danger of becoming too slow for the bonus.. who knows! MrS ____________ Scanning for our furry friends since Jan 2002 | |
ID: 28678 | Rating: 0 | rate: / Reply Quote | |
What will be better...2xGTX690 (4 GPU total) or 2xGTX Titan? I was wondering the same thing and it's confusing for newbies like me when there are statements that refer to Titan's massive compute power as well as statements like "nVIDIA finally got it right with Titan". If I understand correctly, Titan has greater double-precision FP ability than a 690 and if that's true then one certainly can and perhaps should say things like "massive compute power" and "finally got it right". But when you consider that GPUgrid doesn't need double-precision FP then Titan's advantage doesn't mean much if all it's gonna do is crunch GPUgrid tasks. As for exhaust problems on the 690... piece o' cake. I'm going to put four 690s on one mobo if Linux, drivers and BOINC will permit and show y'all how it's done, no fancy case required. Picture 4 - GTX690s, all the same model, nicely lined up in a row, 4 exhaust ports perfectly lined up one above the other, a manifold made from a $6 heat vent boot (unless I can find one in a scrap pile somewhere first) that fits nicely over all four exhaust ports and transitions into 1 collector connected to a suck fan. Once done it's gonna have the highest RAC here for a looooong time. In fact maybe I won't show y'all because then you'll build one too. ____________ BOINC <<--- credit whores, pedants, alien hunters | |
ID: 28680 | Rating: 0 | rate: / Reply Quote | |
And one more point for the Titan: it will be able to crunch GPU-Grid tasks within the bonus time for longer. If the power consumption still allows to run the card when it's in danger of becoming too slow for the bonus.. who knows! The GeForce GTX 285 was released just over 4 years ago. Until the 3.1app was deprecated it could still manage to return tasks for the full bonus. Now using the slower 4.2app (just slower for the 200 series cards) it would depend on the WU. I expect Toni's tasks to return in time, but NOELIA's Long tasks might not. They would still get the 25% bonus though... So probably good for ~4years. If I understand correctly, Titan has greater double-precision FP ability than a 690 and if that's true then one certainly can and perhaps should say things like "massive compute power" and "finally got it right". But when you consider that GPUgrid doesn't need double-precision FP then Titan's advantage doesn't mean much if all it's gonna do is crunch GPUgrid tasks. Exactly. FP64 is not needed here and hasn't been in the past. So it's FP64 Compute benefit isn't applicable to here. I expect the card would still be faster at POEM than a GTX680 or GTX690, but the issue there is the CPU and PCIE over-usage. So you are never going to quite get the most out of the card. It will probably shine at MW, but against the top ATI cards I don't think it's going to be anything special. So it's just an expensive alternative. If an FP64 Fluid Dynamics CUDA based project suddenly appeared then Titan would be the 'bees knees'. ____________ FAQ's HOW TO: - Opt out of Beta Tests - Ask for Help | |
ID: 28694 | Rating: 0 | rate: / Reply Quote | |
Another point I forgot yesterday: Titan can provide more registers per warp and might differ internally in cache sizes and such. I think this has manifested itself in compute performance 2 to 3 times that of a GTX680 in some benchmarks - much higher than the raw horse power implies (Anandtech Compute Bench Part 1 and next page). We can not yet say if anything like this is going to happen at GPU-Grid as well. | |
ID: 28697 | Rating: 0 | rate: / Reply Quote | |
I think they would need to specifically develop towards finding improvements from the registers per warp increase. If it's likely that improvements can be gained from this I'm sure they will try, if they get one to test on. As Titan is CC3.5, any finds could be used in the one app (as it can identify the cards compute capability). The issue I see is the price and limited availability. I can't see a big uptake of the card here, which suggests any such development would be a waste of time, but perhaps lesser versions of the card will appear making it worthwhile. | |
ID: 28699 | Rating: 0 | rate: / Reply Quote | |
We are now trying to get hand on one titan for optimizing the application on it. | |
ID: 28700 | Rating: 0 | rate: / Reply Quote | |
From the benchmarks posted I would expect the Titan's production/$ to be lower than some other solutions (660 TI, 650 TI, etc.). | |
ID: 28705 | Rating: 0 | rate: / Reply Quote | |
What will be better...2xGTX690 (4 GPU total) or 2xGTX Titan? With 2 GTX690 is possible to crunch 4 WU at the same time, with the GTX Titan "only" 2 WU at the same time... ____________ Member of Boinc Italy. | |
ID: 28708 | Rating: 0 | rate: / Reply Quote | |
CUDA Core Count win goes to the 2*690 | |
ID: 28710 | Rating: 0 | rate: / Reply Quote | |
In terms of performance, we are expecting a speed-up of 50% over a gtx680 for normal wu. For large jobs, this could be close to 100% faster. ACEMD reduces the speed when the molecular system is large but not on a titan due to the 6GB of memory. I'm missing a part of the picture here. According to GPU-Z, a GPUGrid job on a GTX 670 uses 384~534MB (depending on the task type) of the 2GB GPU memory. Could you explain please, why would three times more memory on the Titan be more sufficient than the 2GB on the GTX 670/680, when a task is using only the quarter of it? | |
ID: 28712 | Rating: 0 | rate: / Reply Quote | |
acemd uses two types of algorithms, one faster and more memory expensive and one slower that uses less memory. The decision is made at dynamically depending on the gpu memory you have available. | |
ID: 28719 | Rating: 0 | rate: / Reply Quote | |
Tom's Hardware article about TITAN says: As we know, though, Nvidia limits those units to 1/8 clock rates by default—not to be nefarious, but to create more thermal headroom for higher clock rates. That’s why, if you want the card’s full compute potential, you need to toggle a driver switch. Doing this, in my experience so far, basically disables GPU Boost, limiting your games to the card’s base clock rate. Does anyone know what driver switch is that? Has anyone tried that and what would be the result ? | |
ID: 28759 | Rating: 0 | rate: / Reply Quote | |
Does anyone know what driver switch is that? Has anyone tried that and what would be the result ? On Linux systems with K20s, these settings are controlled using the "nvidia-smi" program. Hopefully that will also be so for the Titans. MJH | |
ID: 28761 | Rating: 0 | rate: / Reply Quote | |
Thanks. Does anyone know what driver switch is that? Has anyone tried that and what would be the result ? | |
ID: 28762 | Rating: 0 | rate: / Reply Quote | |
It's a simple driver option in the control panel in windows. It's only being shown if a Titan is present in the system. | |
ID: 28767 | Rating: 0 | rate: / Reply Quote | |
It's a simple driver option in the control panel in windows. It's only being shown if a Titan is present in the system. So that enhancement is disabled in all other consumer cards? | |
ID: 28772 | Rating: 0 | rate: / Reply Quote | |
It's a simple driver option in the control panel in windows. It's only being shown if a Titan is present in the system. The rest of the consumer cards are GK104 (only have low FP64 ability, 1/24th of FP32; 8 FP64 units per SMX block) while Titan is GK110 (up to 1/3rd FP32; 64 FP64 units per SMX block). For Titan, FP64 is set by default to a low level (1/8th speed) and to use FP64 faster you just crank it up. As this is controlled by NVIDIA's System Management Interface (nvidia-smi), apps could turn it up and down. I wonder if this is stepped? I think the GK104 cards just perform FP64 at a speed which increases and decreases with the rest of the GPU. | |
ID: 28775 | Rating: 0 | rate: / Reply Quote | |
Thanks, that makes it clear. | |
ID: 28776 | Rating: 0 | rate: / Reply Quote | |
acemd uses two types of algorithms, one faster and more memory expensive and one slower that uses less memory. The decision is made at dynamically depending on the gpu memory you have available. We already have a rating system for video cards that is broken down into 4 categories: most recommended, highly recommended, recommended and not recommend. So, shouldn't we have the same for video memory size? See example: most recommended : 4 GB + highly recommended : 2 to 4 GB recommended: 1 to 2 GB not recommended : less than 1 GB | |
ID: 28780 | Rating: 0 | rate: / Reply Quote | |
So that enhancement is disabled in all other consumer cards? I'd rather say "this functionality is not present in other consumer cards". MrS ____________ Scanning for our furry friends since Jan 2002 | |
ID: 28784 | Rating: 0 | rate: / Reply Quote | |
We already have a rating system for video cards that is broken down into 4 categories: most recommended, highly recommended, recommended and not recommend. So, shouldn't we have the same for video memory size? That is a good idea. I'll suggest the following classification: - sufficient to run all current tasks at maximum speed - sufficient to run short runs at maximum speed - sufficient to run short runs - insufficient These boundaries would change with apps and WUs and we'd have a hard time figuring them out, so the developers would have to take care of updating this table. MrS ____________ Scanning for our furry friends since Jan 2002 | |
ID: 28785 | Rating: 0 | rate: / Reply Quote | |
acemd uses two types of algorithms, one faster and more memory expensive and one slower that uses less memory. The decision is made at dynamically depending on the gpu memory you have available. I can give you a remote access to my gtx titan if it helps. | |
ID: 28789 | Rating: 0 | rate: / Reply Quote | |
We already have a rating system for video cards that is broken down into 4 categories: most recommended, highly recommended, recommended and not recommend. So, shouldn't we have the same for video memory size? Yikes, it looks like this project might be getting a little rich for my pocketbook, or are you not saying that we should need 4GB cards to do the long runs? Maybe you're not making a 1 to 1 relationship for the 2 classifications above. Right now the long run TONI WUs do OK on the GTX 460 768mb cards (about 12.5 hours). The older long run NATHANS were slow but the last one I received was smaller and ran fine. The 650 TI 1gb card seems to run everything fine currently. | |
ID: 28806 | Rating: 0 | rate: / Reply Quote | |
Sorry if this wasn't clear: Bedrich gave an example of what he had in mind. I proposed different categories, also 4 by accident, totally unrelated to the number he wrote down for his categories. I didn't give any numbers because I don't know them, but I'm sure they'd be smaller than what Bedrich had in mind. | |
ID: 28811 | Rating: 0 | rate: / Reply Quote | |
The analyst and cruncher: | |
ID: 28822 | Rating: 0 | rate: / Reply Quote | |
First test ends with error for short and long 4.2 | |
ID: 28847 | Rating: 0 | rate: / Reply Quote | |
OD !! | |
ID: 28849 | Rating: 0 | rate: / Reply Quote | |
Yes. Maybe application has to be updated for titan. | |
ID: 28858 | Rating: 0 | rate: / Reply Quote | |
It's possible that it does not run until we get one here. | |
ID: 28860 | Rating: 0 | rate: / Reply Quote | |
Ah, -9, my favourite error message. It's an ongoing mystery where this comes from, since there's nothing in the application source that causes a termination with that exit code. | |
ID: 28863 | Rating: 0 | rate: / Reply Quote | |
I have tested 7.0.28, 7.0.52, project reset, no luck. | |
ID: 28866 | Rating: 0 | rate: / Reply Quote | |
First test ends with error for short and long 4.2 Ouch, bleeding edge strikes again. Sorry to hear it and hope there's a resolution soon. | |
ID: 28873 | Rating: 0 | rate: / Reply Quote | |
Installed my titan last night - no luck with gpugrid. | |
ID: 28881 | Rating: 0 | rate: / Reply Quote | |
314.09 is CUDA 5 cappable. | |
ID: 28883 | Rating: 0 | rate: / Reply Quote | |
The Titan card is installed in a machine with a GTX580. I wonder if that's the issue. Can I embed a screenshot in this forum? | |
ID: 28889 | Rating: 0 | rate: / Reply Quote | |
This may be the case. Driver 314.09 support Titan only. Older drivers do not support titan. | |
ID: 28891 | Rating: 0 | rate: / Reply Quote | |
Can I embed a screenshot in this forum? If you don't have your own HTML server or a wqebsite such as most ISPs include with your ISP service, you can upload the image to a free image hosting service like photobucket. They will give you a URL to the image. Then put the URL in a post here. You can embed it by using BBCode tags which are similar to HTML tags. While composing the message find the link that says "Use BBCode tags to format your text" to the left of the edit box and up a little bit to get a summary of what the various tags do. If you can't make the BBCode tags work then just post the URL, we can copy 'n paste it into a browser address bar. ____________ BOINC <<--- credit whores, pedants, alien hunters | |
ID: 28892 | Rating: 0 | rate: / Reply Quote | |
nucleon; | |
ID: 28903 | Rating: 0 | rate: / Reply Quote | |
We need to refresh the app for Titan. Probably get this done on Monday. | |
ID: 28904 | Rating: 0 | rate: / Reply Quote | |
We need to refresh the app for Titan. Probably get this done on Monday. Cheers, I think that's the problem now. It seems a possible bug in gpuz. If I set Titan to be primary, the cuda box is checked, if it's secondary gpu then cuda is deselected. Even though other cuda apps work. Looks like my issue. Oops. -- Craig | |
ID: 28907 | Rating: 0 | rate: / Reply Quote | |
Hi, | |
ID: 28997 | Rating: 0 | rate: / Reply Quote | |
Hi, I think MJH is working on update. | |
ID: 29012 | Rating: 0 | rate: / Reply Quote | |
I installed my new Titan this evening and every single WU is failing within 20 seconds (all NOELIAs). | |
ID: 29023 | Rating: 0 | rate: / Reply Quote | |
I installed my new Titan this evening and every single WU is failing within 20 seconds (all NOELIAs). You all ready knew that was going to happen from you're previous posts and others who rushed out and bought the card. Why didn't you buy a GTX690 instead? It has more cuda cores, costs the same and works right out of the box. | |
ID: 29026 | Rating: 0 | rate: / Reply Quote | |
I thought this card would run cooler than a GTX 690 and didn't want to spend the extra money to water cool a card that will be on it's way out in less than a year. | |
ID: 29027 | Rating: 0 | rate: / Reply Quote | |
Message boards : Graphics cards (GPUs) : GeForce GTX Titan launching the 18th