Show Posts
|
Pages: « 1 2 [3]
|
Latest configurations for both sp20's: How they looking? .worker1
.worker2 (Kill-a-Watt 794w)
![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FqsG9P8t.png%3F1&t=663&c=kpfsE28jOye3tg) Cant seem to get worker1 to be as efficient as 2.
|
|
|
Latest configurations for both sp20's: How they looking? .worker1
.worker2 (Kill-a-Watt 794w)
![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FqsG9P8t.png%3F1&t=663&c=kpfsE28jOye3tg)
|
|
|
Thanks Digitalmocking, I'll do some digging to understand the 'p=' parameter. I'm sure it's been covered on these boards ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) Philip explained it, my settings are so high that I'll only switch over if there's some kind of really high payout available. Interesting idea. Is p=.0155 for a certain amount of hash?
|
|
|
Which LOOPs are front/back? For instance my LOOP 1/3 are running hot (120c) while LOOP 0/2 are fine (85c).
And what is the optimal way to decrease the chip temps? Decreasing volts or watts? Or both?
I understand increasing fans is an option, however I'd like to keep them low as to reduce noise.
EDIT: Assuming the front of the box is where the fan is?
|
|
|
why not match my settings above exactly make sure you have 1 sp20 on 1 evga 1300 g2.
tell us what you get.
in the first 10 minutes
in the first hour
after 1 day
Here's the first hour:![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FqFPT4vp.png%3F1&t=663&c=o5hAEsYD761cOw) Wall A Watt is constant 760W Ambient temps: 13 °C / 60,60 °C Here's after day 1:746W at the wall. ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FxPQfiiM.png%3F1&t=663&c=wcpSEGEwhTLBnA)
|
|
|
Or if you want to save a ton of money over the EVGA's you can use 2880w server Psu's w/ boards, that are more efficient and less than half the price...
Just throwing it out there. For my SP20's I shoot for around 1400-1450 Gh with .67 starting Volts and ~225w per loop. Able to run 3x SP20's per PSU.
Unfortunately already have/purchased the EVGA's. I suppose I could resell them. Just now learning about server grade psu. Do you have a specific brand in mind?
|
|
|
so you are doing 760 with meter and 1300 or 760/1300 = .5846 I am
781/1349 = 0.5789 very close.
that is about what these do at 1300 gh settings.
check your numbers in a ½ day they should stay close to that.
then you can go low.
.631 .631 .631 .631
.641 max
240 watts 240 watts 240 watts 240 watts
reboot at 1000
Perhaps a typo, but did you mean to only lower the volts to .631 and keep the watts the same? Your images show 180 watts, not 240 as you typed above. Also, I just purchased another 1300g2 so I will keep hash higher on both sp20's. Many thanks!
|
|
|
why not match my settings above exactly make sure you have 1 sp20 on 1 evga 1300 g2.
tell us what you get.
in the first 10 minutes
in the first hour
after 1 day
Here's the first hour: ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FqFPT4vp.png%3F1&t=663&c=o5hAEsYD761cOw) Wall A Watt is constant 760W Ambient temps: 13 °C / 60,60 °C
|
|
|
I keep tweaking but can't seem to get above 1200GH and stay under 700W.
I've increased my volts to .66, max .67, watts at 160. Is it better to increase volts or watts at this point?
|
|
|
@philipma1957, or anyone with experience.
At this rate, since I'm using 600W'ish, can I run two SP20's off one (120v) EVGA 1300 G2? Safely? I recall you trying that somewhere in the thread, but don't remember the results, if any?
If so, how do I hook it up if I only have two more VGA red slots off the back?
|
|
|
Here's the pic... >snip image< Can this be better?
theoretically, but each machine has a "character', some better, some a little worse. try this: limit power to each loop to 160w, then play with voltage (min and max)-this would be unique and different for your machine. let machine to adjust iself overnight or at least several hours as initially readings would be slightly lower. For one machine I have: Starting voltage: 0.64 0.64 0.639 0.639 max: 0.649 with fan 10 I get 1225-1231GH at 683w Thanks! Much better (.57/GH). Still tweaking. Trying to get your results, or better. As a side, I noticed the air flow is cooler now that the volts are lower. Here are the current settings/results: ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FIOEccWR.png%3F1&t=663&c=G2JgerDg_aHX8Q) What's the meaning of yellow hz? I read somewhere there's no need for worry, but I'm not sure what hz does/means. Do they need to be white, assuming white it normal?
|
|
|
Here's the pic... ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FVAultXe.png%3F1&t=663&c=J-oCuaFtIzk8iQ) Can this be better? Below is the result after leaving it over night. Temps 16 °C / 61,58 °C ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FTD07WzY.png%3F1&t=663&c=-YynfUZWKd8AMQ)
|
|
|
Here's the pic... ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FVAultXe.png%3F1&t=663&c=J-oCuaFtIzk8iQ) Can this be better?
|
|
|
The new results are somewhere around 1150 hash and 700 watts.
this seems to be a bit low on hash for 700W I agree. Suggestions?
|
|
|
Purchased last week. Just delivered today.
|
|
|
After I posted that, I thought I could do better. The new results are somewhere around 1150 hash and 700 watts. I'll post new pic latter.
Cost is 8.66 cents to 9.33 cents.
I'm still tinkering until getting the most bang for buck. Suggestions welcome.
Also adding another Sp20 but need to get another evga 1300 g2 psu.
|
|
|
Dialed in pretty nicely. For now... ![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FFgkeq9U.png%3F1&t=663&c=NuYuO7s_1Fkahw)
|
|
|
HolyBitcoin delivers! SP20 on my doorstep today.
|
|
|
Just purchased SP20!
Looking forward to hooking it up.
I'm interested to know what the best PSU is for this machine? Is it better to run one or two? What PSU will be most efficient while allowing the sp20 to run at maximum hash rate without pulling tons of wattage. I've read that the EVGA SuperNOVA 1300G2's are good, but pricey too.
Note, It will be plugged in wall outlet, 120v. The electricity cost is .09 per kWh.
|
|
|
|