Wand3r3r feeling hard butt hurt because nobody is buying cheap AMD's crap.
There's no need for posts like this here. -Admin DrPizza
There's no need for posts like this here. -Admin DrPizza
Last edited by a moderator:
Are you serious that lost that you can`t read it from the posts? I even gave you a hint in that thread, which you responded with your usual nonsense. I don`t know if you really are trolling or not but I can post OP`s post again to see if you can make it this time
Now stop wasting my time with your nonsense
This forum frightens me. Its so full of Average Joe`s that is strongly pulled toward AMD that have no idea how things really work in the real world.
Thankfully you're here to educate. How do things work in the "real" world?
Average joes aren't doing CUDA/GPU programming, again, how does the real world affect the average joe such that titan is good performance for the $ or whatever you are trying to promote?
The OP finds some benefit, sure, but you're claiming the average joe is ignorant and uneducated, please enlighten us.
For the PhD minions out there, its time to convince your advisor to buy Titan. We're experiencing a x2 speedup over GTX680 without any code modifications!!! Time to tell your advisor time is money, double the research throughput w/ Titan!!
I don`t think you people understand what OP tries to say. At all.
First of all, lots of professional software are highly CUDA specialized. So OP may not have other choice than Nvidia GPU.
Second, OP is correct about is statement that Titan net him over 2x speed. Titan is over 3.5 faster than GTX 680 on double precision.
Third, he says that time is money, and since Titan is a heck lot faster than 7970, they will pay for the extra cost for a Titan since the Titan can finish a computation much faster than a 7970. With DGEMM (double precision) it can output twice the amount that a 7970 can do.
Probably not.
They must have learnt it from Intel.
Intel loves do the same thing with lowend CPUs, turning features off on purpose.
Anyways title of thread is really misleading.
^ yep, stock 7970 (non ghz ed) beating the 680, by huge amounts.
Also makes you question why in some tests they didnt use any AMD cards to compair.
If there was a card that was "excellent performance per dollar" it would be the 7970.
680 = 133
7970 = 689 (x5 times the 680) (costs ~399$)
7970 1ghz ed = wasnt tested.
Titan = 1309 (x1.8 times 7970) (costs ~999$ = x2.5 of 7970)
71.5 vs 63 = ~13% differnce. (a 7970 1ghz ed would probably be faster)
Price differnce = ~250% differnce.
7970 is a non factor,, unless we get like literally 100 cards for free we will probably never switch to AMD. Our codebase is CUDA and it not economical to switch to OpenCL. We also heavily invested into NVIDIA's libraries (thrust, cublas, etc..) which OpenCL may or may not have similar alternatives.
DP performance actually means nothing in our line of work, we stick to SP.
Actually, we found the performance has high variance, really depends on what you are doing. If your model allows you to fit most of your number crunching in the (smaller) Kepler cache, we notice a significant delta in performance over cache spilling... the delta can be ~30% although we haven't measured too hard on this. This is of course highly dependent on your nature of work.
Wand3r3r feeling hard butt hurt because nobody is buying cheap AMD's crap.
This thread started out with low relevance. It has dropped to irrelevant long ago. We've reverted to the usual empty back and forth.
Well some place they are losing BIG money. Most of it is obviously from the CPUs, but there is no doubt that the GPUs isn`t netting them a whole lot (which I showed) which is why they lost over 1 billion dollars last year despite closing down branches and restructuring the company.
I honestly don`t belive they used much money on R&D either, since the software side is mediocre or bad at best. Money is tight, so they don`t have much money to splash around either. Like I said, they brought this upon themselves.
Look what i just picked up at my local Fry's today: https://www.dropbox.com/s/76k9c2qieh0sq6o/IMG_20130407_150117.jpg
you need better speakers...
These are plenty loud. I'm not an audiophile.
Logitechs are just plain loud. Not that Logitechs are bad, but you'd be surprised how boomy Logitechs are once you listen to some good speakers. Great for games and movies, however.
This thread is just funny. People want Titans, people will buy them. Other people can't leave the first group alone.
One Titan is good enough for me!
This single titan is slower than my Alienware M18x notebook (680m sli with unlocked vbios) that I recently sold. So I definitely need a second, especially at 1440p with AA.
680M = 1344 Stream Processors @ 720MHz with 256-bit memory bus
Titan = 2688 Stream Processors @ 837MHz (876MHz boost) with 384-bit memory bus
Add on top of that SLI scaling and a single Titan is faster.
No it's actually not. I'm going by real world results, not paper specs. The 680m sli was faster in every game I've tested, especially with the unlocked vbios that allowed some heavy oc with no throttling.
680M is a GTX670 with a 27% lower GPU clock (720mhz vs. 980mhz) and a whopping 40% lower memory bandwidth (115 GB/sec vs. 192 GB/sec). That means a 680M is going to be at least 30% slower than a GTX670, or at best only equal to a desktop 660. That's being generous. The Titan at 837mhz is 90-105% faster than GTX660 at 1080P 4-8xAA.
That's only half the story. Once you go up 1 resolution level to Adam's, the 115GB/sec memory bandwidth of low-end cards becomes a liability and even something like a GTX660Ti SLI desktop falls apart. Since 680M SLI is slower than 660Ti SLI, and Titan is 30% faster than HD7970GE at 2560x1440, on average the Titan would beat 680M SLI at 2560x1440 or higher unless you are overclocking the 680M by 20-30%.
Time after time I noticed that you tend to overestimate mobile GPU performance.
Geforce GTX 680M 4GB (DX11) -- 143 VP
Geforce GTX 660 2GB (DX11) -- 161 VP
Geforce GTX TITAN (warmed up, default*) 6GB (DX11) -- 317 VP
Geforce GTX TITAN (temp. boost*) 6GB (DX11) -- 342 VP
http://forums.anandtech.com/showthread.php?t=2300495