hashcat Forum

I have a brand new GPU with 3GB of memory but when I launch the oclHashcat-plus, I can see this output:

Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Workload: 4096 loops, 800 accel
Watchdog: Temperature abort trigger set to 85c
Watchdog: Temperature retain trigger set to 80c
Device #1: Tahiti, 2048MB, 950Mhz, 28MCU
Device #1: Kernel ./kernels/4098/m0400.Tahiti_1124.2_1124.2 (VM).kernel (86896 bytes)
Device #1: Kernel ./kernels/4098/markov_le_plus_v1.Tahiti_1124.2_1124.2 (VM).kernel (88664 bytes)
Device #1: Kernel ./kernels/4098/bzero.Tahiti_1124.2_1124.2 (VM).kernel (30456 bytes)

Other tools like gpu-z report correctly 3GB. Any idea where the problem could be?

I also experience drop of core clock from 950MHz to 850MHz or even more together with drop of load from 99% to 66%. The performance drop does not occur when having the boost button switched off which means the core clock is stable at 850MHz and the load is also stable at 99%. Any idea why this is happening? The attachment shows the unstable case. Thx!

AMD's OpenCL runtime does not allow the use of all memory, see this: https://twitter.com/hashcat/status/373100787108282368

I don't clearly see explanation for my problem in that tweet. Maybe I am a really big newbie. Anyway thx for your reply.

the explanation is that you need to set an environment variable or two to use all available video memory.

Yes, it helped me to increase used memory to 3GB but the speed is still the same. No idea why. Anyway thanks for replying.

it won't affect performance, no. it will only help you e.g. load larger hash lists