The new machine and hard drive are ready. Not sure whether to draw first or wait for compression.
The only item that was downloadable still needed 512gb of memory (more than 466), right now it appears that there are 30 people at the top of a 30ft ladder in the wind.
Where is Harold hidding??? One would hope there would be beta’s to test.
Him simply commenting here in this forum on the current status would go a long way…regardless of what he said. 
This still needs almost 500gb of memory to run.
That’s interesting, I get away with 256 but added few more for stability. But the Windows version is broken by now. Which OS are you using?
Have you changed ramplot to cudaplot?
Hmm, this is disappointing. Not surprising though. CUDA coders are rare as hens teeth. Especially the good ones. They must be an underappreciated bunch.
Are you sure you are running the cuda gpu plotter?
BB ramplot still needs the old 466gb or so.
BB cudaplot only needs 256gb.
I used the first alpha version and it worked perfectly with 256gb
yes,that’s correct.
256gb ram for cuda. but still have bugs.
same one
(ubuntu version)
I think this is just the help files/info has not been updated yet.
If you try running it, it should be fine.
i build version 256/512 buckets
i have 326 GB ram / 1660 cuda 7.5 / 2x 2670v2 /nvme 1tb
I get cuda plot :
./bladebit_cuda -n 1 -c ..... -f ..... cudaplot -t1 /home/ml350g8/dyski/nvme/ /home/ml350g8/dyski/4tb/
result:
Selected cuda device 0 : NVIDIA GeForce GTX 1660
CUDA Compute Capability : 7.5
SM count : 22
Max blocks per SM : 16
Max threads per SM : 1024
Async Engine Count : 3
L2 cache size : 1.50 MB
L2 persist cache max size : 0.00 MB
Stack Size : 1.00 KB
Memory:
Total : 5.80 GB
Free : 5.56 GBAllocating buffers (this may take a few seconds)…
Kernel RAM required : 108297986048 bytes ( 103281.01 MiB or 100.86 GiB )
Intermediate RAM required : 923803648 bytes ( 881.01 MiB or 0.86 GiB )
Host RAM required : 206695301120 bytes ( 197120.00 MiB or 192.50 GiB )
Total Host RAM required : 314993287168 bytes ( 300401.01 MiB or 293.36 GiB )
GPU RAM required : 5494542336 bytes ( 5240.00 MiB or 5.12 GiB )
Allocating buffers
CUDA error: 2 (0x2 ) cudaErrorMemoryAllocation : out of memory*** Panic!!! *** Fatal Error:
CUDA error cudaErrorMemoryAllocation : out of memory.
./bladebit_cuda(+0xe09c5)[0x55f5b86719c5]
./bladebit_cuda(+0xe019f)[0x55f5b867119f]
./bladebit_cuda(+0x2190d)[0x55f5b85b290d]
./bladebit_cuda(+0x21ecb)[0x55f5b85b2ecb]
./bladebit_cuda(+0x188bb)[0x55f5b85a98bb]
/lib/x86_64-linux-gnu/libc.so.6(+0x29d90)[0x7f88e9ee9d90]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x80)[0x7f88e9ee9e40]
./bladebit_cuda(+0x19eb5)[0x55f5b85aaeb5]
How much is required ram ,also how to I getting mistake ?
Nvidia smi:±--------------------------------------------------------------------------------------+
| NVIDIA-SMI 530.30.02 Driver Version: 530.30.02 CUDA Version: 12.1 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA GeForce GTX 1660 On | 00000000:21:00.0 Off | N/A |
| 0% 35C P8 8W / 130W| 172MiB / 6144MiB | 4% Default |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+
+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| 0 N/A N/A 2161 G /usr/lib/xorg/Xorg 162MiB |
| 0 N/A N/A 4324 G ...gnu/webkit2gtk-4.0/WebKitWebProcess 7MiB |
+---------------------------------------------------------------------------------------+
and memory test 
ml350g8@ml350g8:~/bladebit/build-release$ ./bladebit_cuda --memory
required : 446676598784
total : 354979307520
available: 334612115456
ml350g8@ml350g8:~/bladebit/build-release$
You are running out of Vram i think, not system ram
Minimum req is 8GB vram
I have a 12gb video card, and still crashes
mistake
look this threads
---------->-------------------->============>
Quote :
GTX 1060 6GB (hobbled on a 8x slot, no less) benchmarks at 11.3 minutes. This really needs to be incorporated into the mainline.
Will the 1060 work and the 1660 won’t?
1660 and 1060 both have same problem, not enough memory.
The users in the github thread made a (small) alteration in the code and then build their own version to reduce the memory requirements.
It is possible the official release will support 6GB, but so far Harold has been sticking with 8GB minimum. so we’ll have to wait and see I suppose.
I also built according to the instructions. The question is why do I want more than 400 gb of ram if there is info about 256 on git
and the problem starts with loading into ram
For my build is requried:
Allocating buffers (this may take a few seconds)...
Kernel RAM required : 92836995072 bytes ( 88536.26 MiB or 86.46 GiB )
Intermediate RAM required : 1568940032 bytes ( 1496.26 MiB or 1.46 GiB )
Host RAM required : 176093659136 bytes ( 167936.00 MiB or 164.00 GiB )
Total Host RAM required : 268930654208 bytes ( 256472.26 MiB or 250.46 GiB )
GPU RAM required : 5582622720 bytes ( 5324.00 MiB or 5.20 GiB )
Allocating buffers
I have :
RAM
331GB
Card Memory :
Memory:
Total : 5.80 GB
Free : 5.63 GB
Plotting buffored 262GB ram and stopped
Why ??
Hope it actually comes out in the second quarter, or at least before the end of the year !