Recent content by ahu

  1. A

    CryptoCurrency Mining with GPUs *spawn*

    At least according to this the single GPU Titans have a history of keeping their value. And I don't really see this changing with this model. This one being a real GPGPU workhorse suitable for AI and HPC workloads. Come two years I'd MUCH rather have a couple of Titan V's rather than a pile of...
  2. A

    CryptoCurrency Mining with GPUs *spawn*

    Are you referring to this? If this is true, the slowdown is way bigger than expected. From what I heard, Meltdown mainly slows code that frequently changes between user mode and kernel mode. And I thought the there was nothing of the sort in CryptoNote.
  3. A

    CryptoCurrency Mining with GPUs *spawn*

    I've now started to mine with my Titan V. Currently mining Zcash with it, along with CPU mining Monero on the Xeons. Getting 900 Sol/s with stock clocks. Contrary to what most people on the forums seem to say, I find Titan V an excellent choice for mining if you can afford it. Because not only...
  4. A

    CryptoCurrency Mining with GPUs *spawn*

    I've tried that, but after 20 threads the performance deteriorates whatever I do. The best I could get with HT on was around 850 H/s, and that was only with the real even numbered cores. With HT turned off, I can get 960 H/s with "low_power_mode" : false on all cores.
  5. A

    CryptoCurrency Mining with GPUs *spawn*

    That's interesting. I just came back to CPU mining after a hiatus of 2 years and have the exact same Xeon as you, two of them actually. Do you have hyperthreading disabled? With my system I was doing close to 900 H/s with 20 cores and both hyperthreading and low power mode disabled. Looking...
  6. A

    NVIDIA Kepler speculation thread

    A new driver won't help much here, as the integer performance is severely handicapped compared to GTX 580 and even to GTX 560. According to CUDA C Programming Guide version 4.2, 32-bit integer shifts and compares have only 1/24 of the throughput of the 32-bit FMA. That would put the GTX 680 at...
  7. A

    Process Explorer 15 now supports GPU

    It's impressive to see all the 48 GPU engine graphs on a 8 GPU system consisting of three Nvidia and one ATI dual-GPU cards :cool: There's obviously room for improvement though, because I can't even directly see which engine belongs to which GPU. While Russinovich is saying the GPU driver is...
  8. A

    NVIDIA Maxwell Speculation Thread

    64-bit confirmed for Project Denver CPU :smile: Preliminary silicon shown here: http://www.brightsideofnews.com/news/2011/3/9/nvidia-reveals-64-bit-project-denver-cpu-silicon-die.aspx The FPU portion seems rather large to me.
  9. A

    The AMD Execution Thread [2007 - 2017]

    <offtopic> That's peanuts. I've actually seen a 100 trillon dollar note :twisted: It was Zimbabwean dollars though and worth some 50 american dollars. </offtopic>
  10. A

    NVIDIA Maxwell Speculation Thread

    Agreed. And two more points: - Fermi *already* has 64-bit addressing - Maxwell and Project Denver are supposed to be used in supercomputers So I would expect Maxwell ARM core to be at least 64-bit.
  11. A

    NVIDIA Maxwell Speculation Thread

    Well addressing single objects larger than 4 GB for instance. And are you suggesting that being forced to use paging methods would be OK from year 2014 onwards?
  12. A

    NVIDIA Maxwell Speculation Thread

    In a related and confusing news, ARM doesn't have any immediate plans for 64-bit architecture: http://www.xbitlabs.com/news/cpu/display/20110204225101_ARM_Has_No_Immediate_Plans_for_64_Bit_Chips_Chief_Exec.html So, will Project Denver incorporate the first 64-bit ARM-core? I really can't...
  13. A

    AMD: R9xx Speculation

    I'm sorry but I only see GTX580 SLI beating even the overclocked 6970 CrossFireX in the link. But it's true though that AMD has been getting better with CrossFire lately, and they are now at least on par now with Nvidia on scaling. I'm now looking forward to the 6990 model. The combination...
  14. A

    AMD: R9xx Speculation

    Your comparison is a bit flawed as GTX580 is a far more powerful card and thus more prone to system bottlenecks. From the same link: GeForce GTX 460 1024MB: 79%
  15. A

    NVIDIA Fermi: Architecture discussion

    Where did you get that? The point system tends to reflect the actual scientific value for given workload. You might say it's "unfair" only because it doesn't match the GFLOPS output. But that's only a result of different optimization. On AMD you have to do certain calculations twice, whereas...
Back
Top