[QUOTE=DrobinsonPE;554782]
here is the output of ls -l from the folder mfaktc/mfaktc021

-rw-rw-rw- 1 user user 18845 Feb 17 2015 Changelog.txt
-rw-rw-rw- 1 user user 35147 Feb 17 2015 COPYING
-rwxrwxr-x 1 user user 2719952 Aug 22 20:40 mfaktc.exe
drwxrwxr-x 2 user user 4096 Sep 4 2019 src
-rw-rw-r- - 1 user user 55 Aug 22 20:55 worktodo.txt
[/QUOTE]No mfaktc.ini?!

[QUOTE=kriesel;554787]No mfaktc.ini?![/QUOTE]

Linux can be such a pain in the keister. Does the Linux version of [I]mfaktc[/I] not have [I]mfaktc.ini[/I] in its archive? I thought it did...

[QUOTE=kriesel;554787]No mfaktc.ini?![/QUOTE]

You caught that. Yes, I realized that after the fact as well and that was a stupid mistake I made out of frustration on the 22nd. When the new compiled mfaktc did not work, I started to delete the new folder (mfaktc021) I had created and then stopped. I must not have pulled the .ini file back out of the trash. That did compromise the testing on the 23rd.

However, mfaktc does now work with the .ini file and the worktodo file located in my user folder (also called Home in Linux Mint)

I think the key is that mfaktc.ini and worktodo.txt have to be in the folder where you run the command. Which does not have to be the same place as the command itself.

So if you want to use a command for more than one task you can set up a folder for each task with appropriate mfaktc.ini and worktodo.txt (or whatever the command you are using needs) in them, then run [c]/full/path/to/command[/c] in each directory.

Sometimes learing Linux is easier if you have not had much experience with Windows.

Chris

[QUOTE=chris2be8;554833]..Sometimes learing Linux is easier if you have not had much experience with Windows.

Chris[/QUOTE]

Add MS-DOS. I started with this in 1988. Add nearly every Windows version there ever was. There you have it. My Linux learning experience was not so bad because I started with [I]Ubuntu[/I], which is mostly GUI. The console part took a little longer. I learned what I needed to get by and maybe a little more.

[U]Q[/U].: I have seen references to [I]mfaktc v0.22[/I]. What is the story with it?

[QUOTE=chris2be8;554833]I think the key is that mfaktc.ini and worktodo.txt have to be in the folder where you run the command. Which does not have to be the same place as the command itself.[/QUOTE]

It took me a while to figure out what you said, but I get it now. Thank you for the explanation.

I have been running my GPU for a few weeks now and thought I would share some results. Mfaktc and mprime run very slightly faster when they are running together. Power usage is, mprime alone 73W, mfaktc alone 145W, mprime and mfaktc 182W. The computer is an i3-4130T with a GTX-1650 Super. Most of the time I run the GPU at nvidia-smi -PL 70. The same effect happens at -PL70 but it is a smaller difference.

[CODE]mprime 30.3 build 4 and mfaktc-0.21 running at nvidia-smi -PL100

mfaktc before mprime:
M98413771 from 2^76 to 2^77 [mfaktc 0.21 barrett87_mul32_gs]
Sep 05 08:00 | 288 6.4% | 10.793 2h41m | 1296.74 82485 n.a.%
Sep 05 08:00 | 293 6.5% | 10.818 2h41m | 1293.75 82485 n.a.%

mfaktc with mprime:
M98413771 from 2^76 to 2^77 [mfaktc 0.21 barrett87_mul32_gs]
Sep 05 08:13 | 641 14.0% | 10.774 2h28m | 1299.03 82485 n.a.%
Sep 05 08:13 | 644 14.1% | 10.775 2h28m | 1298.91 82485 n.a.%
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| 0 GeForce GTX 165... Off | 00000000:01:00.0 Off | N/A |
| 63% 76C P2 98W / 100W | 460MiB / 3909MiB | 100% Default |
+-------------------------------+----------------------+----------------------+

mprime with mfaktc:
[Work thread Sep 5 08:26] Iteration: 490000 / 54082711 [0.90%], ms/iter: 9.066, ETA: 5d 14:57
[Work thread Sep 5 08:28] Iteration: 500000 / 54082711 [0.92%], ms/iter: 9.062, ETA: 5d 14:52

mfaktc after mprime:
Sep 05 08:34 | 1184 25.8% | 10.796 2h08m | 1296.38 82485 n.a.%
Sep 05 08:34 | 1185 25.9% | 10.797 2h07m | 1296.26 82485 n.a.%

mprime running alone:
[Work thread Sep 5 21:31] Iteration: 700000 / 54082711 [1.29%], ms/iter: 9.128, ETA: 5d 15:20
[Work thread Sep 5 21:32] Iteration: 710000 / 54082711 [1.31%], ms/iter: 9.126, ETA: 5d 15:18[/CODE]

[QUOTE=DrobinsonPE;556201]Most of the time I run the GPU at nvidia-smi -PL 70.[/QUOTE]

Out of curiosity, what is your power consumption and thruput if you lock the graphics clock to base clock ([C]nvidia-smi -lgc 1530[/C])?

You can reset graphics clock by [C]nvidia-smi -rgc[/C] afterwards.

[QUOTE=axn;556217]Out of curiosity, what is your power consumption and thruput if you lock the graphics clock to base clock ([C]nvidia-smi -lgc 1530[/C])?[/QUOTE]

I like these results. It lowered the power usage and increased the efficiency. I was trying to get my power use down to this level but it would not go below -PL70.

[CODE]mfaktc with PL-70, Power usage 111W:
Sep 06 07:23 | 1677 36.5% | 11.591 1h57m | 1207.16 82485 n.a.%
Sep 06 07:23 | 1680 36.6% | 11.627 1h58m | 1203.42 82485 n.a.%
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| 0 GeForce GTX 165... Off | 00000000:01:00.0 Off | N/A |
| 52% 66C P2 66W / 70W | 477MiB / 3909MiB | 100% Default |
+-------------------------------+----------------------+----------------------+

mfaktc with PL-70 and nvidia-smi -lgc 1530, Power usage 92W:
Sep 06 07:33 | 1884 40.7% | 12.707 2h00m | 1101.14 82485 n.a.%
Sep 06 07:33 | 1889 40.8% | 12.713 2h00m | 1100.62 82485 n.a.%
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| 0 GeForce GTX 165... Off | 00000000:01:00.0 Off | N/A |
| 43% 58C P2 53W / 70W | 488MiB / 3909MiB | 100% Default |
+-------------------------------+----------------------+----------------------+

mfaktc with PL-100 and nvidia-smi -lgc 1530, Power usage 92W:
Sep 06 07:42 | 2057 44.7% | 12.710 1h52m | 1100.88 82485 n.a.%
Sep 06 07:42 | 2060 44.8% | 12.707 1h52m | 1101.14 82485 n.a.%
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| 0 GeForce GTX 165... Off | 00000000:01:00.0 Off | N/A |
| 43% 59C P2 54W / 100W | 488MiB / 3909MiB | 100% Default |
+-------------------------------+----------------------+----------------------+[/CODE]

[QUOTE]Originally Posted by [B]DrobinsonPE[/B] [URL="https://www.mersenneforum.org/showthread.php?p=554782#post554782"][IMG]https://www.mersenneforum.org/images/buttons/viewpost.gif[/IMG][/URL]
[I]here is the output of ls -l from the folder mfaktc/mfaktc021

-rw-rw-rw- [COLOR=Red]1[/COLOR] user user 18845 Feb 17 2015 Changelog.txt
-rw-rw-rw- [COLOR=red]1[/COLOR] user user 35147 Feb 17 2015 COPYING
-rwxrwxr-x [COLOR=red]1[/COLOR] user user 2719952 Aug 22 20:40 mfaktc.exe
drwxrwxr-x [COLOR=red]2[/COLOR] user user 4096 Sep 4 2019 src
[/I][I] -rw-rw-r- - [COLOR=red]1[/COLOR] user user 55 Aug 22 20:55 worktodo.txt[/I][/QUOTE]What is the significance of the numbers I have marked in red? I have two which are zero. This resulted in hidden files. Is there any way to change them?

[QUOTE=storm5510;557069]What is the significance of the numbers I have marked in red? I have two which are zero.[/QUOTE]

This is your "inode" link count.

[QUOTE=storm5510;557069][FONT="Courier New"]-rwxrwxr-x 1 user user 2719952 Aug 22 20:40 mfaktc.exe[/FONT][/QUOTE]

That's weird. Why does it have an .exe file extension? That's usually not the case in Linux.
And while you can omit the .exe extension in CMD, that's not valid for BASH etc. So if you have your mfaktc named with extension, you'll have to write [c]./mfaktc.exe[/c].