75-76_253-297M taking 4 minutes 19 seconds 14,000 credits

This post is slightly different to my normal ones in this thread.

The two tasks below that I have also included the link for at the top of the writing I also ran manually I noticed for the bottom task that last number is 257-387M I believe should run faster than the top one since it is lower credit. However it is actually 3.66 GHz days more than the top one. Is there a particular reason why my bottom task has a longer runtime?
I am now using a 4090 limited to 320W

https://srbase.my-firewall.org/sr5/result.php?resultid=26057509 gave 12,000 credit for 2 minutes 36 seconds (runtime)
got assignment: exp=376317883 bit_min=75 bit_max=76 (20.33 GHz-days)
Starting trial factoring M376317883 from 2^75 to 2^76 (20.33 GHz-days)
k_min = 50195504343600
k_max = 100391008691333
Using GPU kernel "barrett76_mul32_gs"
Date Time | class Pct | time ETA | GHz-d/day Sieve Wait
Jun 09 10:19 | 4617 100.0% | 0.159 n.a. | 11509.87 82485 n.a.%
no factor for M376317883 from 2^75 to 2^76 [mfaktc 0.23.0 barrett76_mul32_gs]
tf(): total time spent: 2m 33.168s

https://srbase.my-firewall.org/sr5/result.php?resultid=25451631 gave 11,700 credits for 3 minutes 2 seconds (runtime)
got assignment: exp=318912029 bit_min=75 bit_max=76 (23.99 GHz-days)
Starting trial factoring M318912029 from 2^75 to 2^76 (23.99 GHz-days)
k_min = 59230960930680
k_max = 118461921870489
Using GPU kernel "barrett76_mul32_gs"
Date Time | class Pct | time ETA | GHz-d/day Sieve Wait
Jun 09 10:22 | 4615 100.0% | 0.188 n.a. | 11486.65 82485 n.a.%
no factor for M318912029 from 2^75 to 2^76 [mfaktc 0.23.0 barrett76_mul32_gs]
tf(): total time spent: 3m 0.423s

Things have returned to normal tasks that are giving 11,700 credits are taking 2 minutes 37 seconds and they have dropped to just over 20 GHz days
https://srbase.my-firewall.org/sr5/result.php?resultid=25456175

Meant to ask in my original post has the number of tasks per bit level changed I seem to recall there were 100,000 work units before the range would change. How is this changed to close to 200,000?

Thank you. I must've been unlucky to receive almost a full cache around 75 tasks that ran for a touch over 3 minutes.
I know we are not halfway through the current bit level yet, is it too early to ask whether we will be processing another bit level when this one is completed?

No it's not too small. I was simply curious

My 4090 takes around 3 hours 26 minutes/2 minutes 36 seconds a task to complete 80 tasks, bearing in mind it is limited to 350W. If it's going to take 1 to 2 years to complete the millions of tests remaining they could get too quick for my GPU. That's not a bad thing I guess

Thanks for the details

So I am able to have a rough idea of tracking progress by looking at the 5 numbers at the end of the task.

Taking 75-76_293-387M_wu_29659 is it reasonable for me to work out the progress with task number "29659" that we are 29.65% of the way through 293 – 387M range?

For those interested my 4090 at 320W took 2 minutes 35 seconds to complete task

Thanks, I am unsure what to do with this information, in relation to the question I asked

Thank you, my maths is on point in regards to task numbers

75-76_360-392M taking 2 minutes 38 seconds 11,700 credits

we are nearly at 400k range.

Appreciate the confirmation. I do not recall credit dropping by 1200 previously when changing to a high bit level less credit, less runtime.

Appreciate the confirmation. I do not recall credit dropping by 1200 previously when changing to a high bit level less credit, less runtime.

It will be reduced from time to time after decreasing range too.