Topic: This is a clock rate to death rate thread. For more current gear (Read 1068 times)
anything >9000
March 28, 2015, 07:33:03 PM
Dragon miner (4 blade) from pfcli
1300 evga g2
1 blade dead within 7 days
2nd blade catches fire (spectacular) within 60 days
ambient temp 22 deg celcius
1300 evga g2
1 blade dead within 7 days
2nd blade catches fire (spectacular) within 60 days
ambient temp 22 deg celcius
March 28, 2015, 06:56:58 PM
LKTC Dragon 1T
Operating in server room of 68F.
Stock Settings.
After 4 months, dead stock PSU, replaced with EVGA 1300 G2
After 7 months, one blade quit hashing.
Bitmain S5
After approximately 2 weeks of stock hashing in same 68F server room,
one board goes dead.
Hard reset and rinse & repeat every 2 weeks (within days).
Corsair TX850 PSU.
Operating in server room of 68F.
Stock Settings.
After 4 months, dead stock PSU, replaced with EVGA 1300 G2
After 7 months, one blade quit hashing.
Bitmain S5
After approximately 2 weeks of stock hashing in same 68F server room,
one board goes dead.
Hard reset and rinse & repeat every 2 weeks (within days).
Corsair TX850 PSU.
Had a Lketc dragon PSU completely crap out. It was during winter I would say under 50 degrees everyday. I did have it since first released though... so it stuck with me a long time.
Never OC'ed just ran stock settings.
Never OC'ed just ran stock settings.
What psu?
was the room at 50f
what did internal temps read?
the winter 150 days or 90 days .
And the psu failed but a replacement psu works?
Your gear would be psu failure.
More data. please
Good idea for a thread!
I've never had an ASIC fail so far, but if you're interested I'd be happy to post the speeds I've pushed them to without issue.
Frequency doesn't kill chips, voltage and heat does. The only way frequency can damage hardware is if the power delivery is garbage and already close to it's limits at stock frequency, as is the case with KNC's junk. That would lead to dead VRMs, etc, but the chip wouldn't be the weak link. Shitty power supplies with bad ripple suppression are also hardware killers.
I've never had an ASIC fail so far, but if you're interested I'd be happy to post the speeds I've pushed them to without issue.
Frequency doesn't kill chips, voltage and heat does. The only way frequency can damage hardware is if the power delivery is garbage and already close to it's limits at stock frequency, as is the case with KNC's junk. That would lead to dead VRMs, etc, but the chip wouldn't be the weak link. Shitty power supplies with bad ripple suppression are also hardware killers.
give me details on your success.
this will need a lot of data to come close to show any failure numbers with any shot of success.
Like I ran 2 s-5's at freq 393 and 400 with good cooling for a month zero failure. Sounds like enough info. No it is not.
I am missing
psu used
room temp
asic temp
pool used all missing
Just for clarification, only those listed mfg?
Well do any asic but for btc .
Had a Lketc dragon PSU completely crap out. It was during winter I would say under 50 degrees everyday. I did have it since first released though... so it stuck with me a long time.
Never OC'ed just ran stock settings.
Never OC'ed just ran stock settings.
Good idea for a thread!
I've never had an ASIC fail so far, but if you're interested I'd be happy to post the speeds I've pushed them to without issue.
Frequency doesn't kill chips, voltage and heat does. The only way frequency can damage hardware is if the power delivery is garbage and already close to it's limits at stock frequency, as is the case with KNC's junk. That would lead to dead VRMs, etc, but the chip wouldn't be the weak link. Shitty power supplies with bad ripple suppression are also hardware killers.
I've never had an ASIC fail so far, but if you're interested I'd be happy to post the speeds I've pushed them to without issue.
Frequency doesn't kill chips, voltage and heat does. The only way frequency can damage hardware is if the power delivery is garbage and already close to it's limits at stock frequency, as is the case with KNC's junk. That would lead to dead VRMs, etc, but the chip wouldn't be the weak link. Shitty power supplies with bad ripple suppression are also hardware killers.
Just for clarification, only those listed mfg?
space post
space
Simple thread.
old info deleting it.
old info deleting it.
Jump to: