 Show Posts
|
|
Pages: [1] 2 3 4 »
|
i am advertising S1 units localy running at 140gh/s 165watt at 750$ including psu and getting some bites.
Got less interest when i advertised them at 180 or 200 with the high watt usage.
Well that is not really honest of you, as the specs of the BM1380 chip does not really support your Gh/W claim! I have one of my S1 running 0.86V @250Mhz and it draws 164W at the wall. So I really would like you to show me how to get it to 140Gh/s without using more power!!! okay and at your settings of 250mhz and pulling 164 watts at the wall what is the hash? 128gh? 135gh? also what is the psu? as a really good plat psu saves 5% and my 2 orders now marked paid. Sry, yes: 128Gh and the PSU: http://www.xfxforce.com/en-us/products/xtr-series-full-modular/xtr-series-650w-psu-p1-650b-befxokay I know that psu. if you used this one http://www.newegg.com/Product/Product.aspx?Item=N82E16817371074you would run a bit lower. maybe 128gh and 155 watts. also s-1s vary I have 3 over clocked to 400freq and they all use a different amount of power and 2 hash at 204gh and one hashes at 195gh. the point is maybe 140gh at 165 watts could be done. maybe not. For the one that is running 195GH, have you tried the 393 setting? These settings worked like a charm on one of my S1 that gace HW error galore on 400Mhz: #option 'freq_value' '5f05' #393.75M #option 'chip_freq' '393.75' #option 'timeout' '34' |
|
|
|
i am advertising S1 units localy running at 140gh/s 165watt at 750$ including psu and getting some bites.
Got less interest when i advertised them at 180 or 200 with the high watt usage.
Well that is not really honest of you, as the specs of the BM1380 chip does not really support your Gh/W claim! I have one of my S1 running 0.86V @250Mhz and it draws 164W at the wall. So I really would like you to show me how to get it to 140Gh/s without using more power!!! okay and at your settings of 250mhz and pulling 164 watts at the wall what is the hash? 128gh? 135gh? also what is the psu? as a really good plat psu saves 5% and my 2 orders now marked paid. Sry, yes: 128Gh and the PSU: http://www.xfxforce.com/en-us/products/xtr-series-full-modular/xtr-series-650w-psu-p1-650b-befx |
|
|
|
i am advertising S1 units localy running at 140gh/s 165watt at 750$ including psu and getting some bites.
Got less interest when i advertised them at 180 or 200 with the high watt usage.
Well that is not really honest of you, as the specs of the BM1380 chip does not really support your Gh/W claim! I have one of my S1 running 0.86V @250Mhz and it draws 164W at the wall. So I really would like you to show me how to get it to 140Gh/s without using more power!!! |
|
|
|
Zero. BFL has nothing to burn in, ergo there can't be any hashrate from burn-in.
Ok, from preorders then...? |
|
|
|
I would describe the noise as something from a quiet vacuum cleaner.
That's kinda subjective! A real measurement in dB would be nice, a video of the measurement even better!! |
|
|
|
Any photos, video...?
I'm sorry but I didn't have my camera with me ( stupid me forgot it ), but Noncetech uploaded 2 videos to their web site last night. You don't own a phone? Please cyberspaced no need to be snide! Virtualspade was extremely nice to visit our demo event personally. Even more nice of him to comment on what he saw. There's plenty of footage; both videos and images available of the mining performance of our product in the launch thread: https://bitcointalk.org/index.php?topic=562182.0We can provide you with any information you feel that is missing if you contact us directly via email: info@noncetech.comVideos and a few photos from one source...! You guys moved from Finland to China, so my guess is that most people here would like to see some proof from someone not affiliated with Noncetech. And why can't you provide info here, could end up saving you answering a lot of mails... If possible, can you tell me the noise level in dB of one card running turbo ( they sound really loud in your video )? |
|
|
|
Any photos, video...?
I'm sorry but I didn't have my camera with me ( stupid me forgot it ), but Noncetech uploaded 2 videos to their web site last night. You don't own a phone? |
|
|
|
I went to check out Noncetech's beta miners in Helsinki today, generally speaking I liked what I saw, I admit I'm not a Bitcoin miner expert but I have some general knowledge about electronics and Bitcoin mining. I started mining scrypt coins about a year ago and I have read a lot of articles about Bitcon mining and mining hardware in the last year. I would say I have pretty good picture about Bitcoin mining in general.
The whole team was there, the coder, hardware guy and the economics guy, everyone seemed to know their stuff pretty well. Also one of the guys has a Chinese origin and speaks Chinese which I would say is a good advantage when you do business with Chinese subcontractors, for the assembly of the miners and so on. All the guys were friendly and in my opinion "down to earth" and none of them were full of themselves. Well that may yet change when they get filthy rich selling all those miners to us, I noticed money can do strange things to good people, I just hope they get rich, and don't turn into the dark side after that... I liked their team and wouldn't mind doing business with them.
They set up 2 miners at the occasion, one was their first test miner the other was one of the "beta" miners. The hardware setup seemed quite basic, just plug in the cables and boot the RasPi, the software side was not not yet that simple, their drivers for cgminer was not yet completely finished, so they needed to manually input some commands ( setting the voltage to the chips ) to the RasPi before starting cgminer.
Their web interface for the miner was not yet finished, so I just got a look at the page layout for the interface. The driver part that controls the fan speed on the miner was not yet fully programmed so the fans were on full speed all the time, and those small fans are quite noisy.
The miners was running in "Turbo" mode hashing around 250GH/s or more correctly they showed in cgminer as 2 x 125GH/s per card, 2 hashing blocks per card. One miner showed a little more HW errors than the others, that was the original first test miner the built, on the "Beta" miners they had made some hardware modifications on the Pcb, and it had less HW errors or none. We calculated the percentage of the HW errors on the original test miner, but stupid me cant remember the exact number, but it was low. As a scrypt miner with Gpu's I'm used to 0 HW errors, if I have HW errors I re tune my rigs so they go away, I was assured that Bitcoin mining with Asic's will produce some HW errors and thats normal.
The components on the Pcb's are of high quality, all critical components are made in USA or Japan, this is for stability and long life of the miner, I agree with the choice for a little more expensive components, as for me a miner has to be production grade, I don't have time to baby sit my miners, they should work 24/7/365.
One thing that I didn't get to see in the 3h visit I made to Noncetech is the chassis for the miners, they have not yet got that from the producer. Well I of course wonder how hot will the chassis get on the inside, with 4 cards running in "Turbo" mode producing 1000w of heat, and 2x120mm fans to get that heat out from there.
I'm no expert, but I don't see any major issues with Noncetech getting their miner shipped within a reasonable time after the Pre-orders start, there is some coding to do, but I would say that the coding is not a big issue. I was told that they have the "assembly line" all ready arranged with a Chinese subcontractor, and they are going to China to oversee the assembly and shipping of the miners. One thing that comes to mind is the chassis, if it needs total redesign, that could cause a small delay. Another thing that they have not done yet is a long term stress test, running a miner more than a day or two, now that I think of it, I should have volunteered to do a 3 week stress test on some of their miners, I'm sure my scrypt rigs wouldn't mind the noise in the garage...
Ok, I wont bore you guys any more today..
Best of luck to Noncetech launch today
Any photos, video...? |
|
|
|
I would really like to know how you get a S1 to 150GH using only 150 Watt!! There is a thread with a lot of info on how to undervolt and underclock an s-1. Right now diff and power do not call for the mod. https://bitcointalk.org/index.php?topic=526060.0 this thread describes a how to that achieved 140gh and 160 watts. [/quote] If you have a look at the specs of the BM chip you will see that thoose numbers does not add up, @140GH it will use +180W! I have one running 0.86V 250M, it gives 128GH and 164W at the wall. |
|
|
|
so fast questions. who has one mining at btcguild ? who has one getting 1th ? who has one at 950 watts or less? So far from what I read I am not that impressed. Seems to me a 2x s-1 under clocked to 300 gh pulling 300 watts would have been: A easier to make B cheaper to ship C fit in a much smaller case Frankly I think they fucked up. They will need a 1200 watt psu or sell it with out one . Here is a prediction buying these psu's at amazon will allow for one to be marked up to 400 from 290. http://www.amazon.com/gp/product/B00JG5LSN8/ref=oh_details_o00_s00_i00?ie=UTF8&psc=1 this psu is basically the same as a corsair ax1200i . after reading the thread i ordered one of these . I already see these marked way the fuck up I would really like to know how you get a S1 to 150GH using only 150 Watt!! |
|
|
|
Anyone managed to push theirs past 225?
Every time I 250, it jumps up to 1275W in, but beeps constantly and eventually drops back down in 5-10 seconds.
Edit: At 225M I get 1153GH/s using 1148W using an EVGA 1300W G2 on 115V. It will probably be a little better on 240V, but I don't have per-outlet metering on my PDU and I wanted to get power measurements using my KillAWatt.
To get it running @ +225Mhz I would suggest to up the voltage to 0.85V... Get out your pencil!! Where's the rocket scientist when u need one? check the overclocking thread. There will be two resistors associated with the output voltage of the TPS53355 regulator. using a pencil mod on one will increase voltage and pencil modding the other will decrease it. Figure out which one ups the voltage and go to town - 0.9V will probably unlock ~1.25TH/1350W. 0.95V might be sufficient for up to ~1.35TH/1.6kW if you can cool it a lot better then stock I have a S1 on 0.86V and max clock is 250M (128GH), with 164W at the wall. 275M gives 138GH but over 10% HW error and 180W consumption. sounds about right then - the S2 is basically 10 S1 units running at clocks similar to your current ones (ps: why dont you run your S1 at stock clocks right now - You would be making a lot more profit then with your underclocked speeds) A small pencil mod to increase to S2 voltage should make >1.25TH possible. The PCI slot seems to use mutliple pins on a common blade of the cards to provide voltage. It looks pretty good method, and it *might* work with a ribbon cable, but a fat pair of 12AWG wires to carry the voltage would be the best way MrTeal - what sort of temperatures does it hit at 225MHz - did you modify cooling? I have more than one S1. I ghettoed one just to check performance and durability of the pencil mod. It runs at 41-43 degrees C ( room temp. 20C ), 1320RPM fanspeed and 3 HW errors a day in average! |
|
|
|
Anyone managed to push theirs past 225?
Every time I 250, it jumps up to 1275W in, but beeps constantly and eventually drops back down in 5-10 seconds.
Edit: At 225M I get 1153GH/s using 1148W using an EVGA 1300W G2 on 115V. It will probably be a little better on 240V, but I don't have per-outlet metering on my PDU and I wanted to get power measurements using my KillAWatt.
To get it running @ +225Mhz I would suggest to up the voltage to 0.85V... Get out your pencil!! Where's the rocket scientist when u need one? check the overclocking thread. There will be two resistors associated with the output voltage of the TPS53355 regulator. using a pencil mod on one will increase voltage and pencil modding the other will decrease it. Figure out which one ups the voltage and go to town - 0.9V will probably unlock ~1.25TH/1350W. 0.95V might be sufficient for up to ~1.35TH/1.6kW if you can cool it a lot better then stock I have a S1 on 0.86V and max clock is 250M (128GH), with 164W at the wall. 275M gives 138GH but over 10% HW error and 180W consumption. |
|
|
|
Anyone managed to push theirs past 225?
Every time I 250, it jumps up to 1275W in, but beeps constantly and eventually drops back down in 5-10 seconds.
Edit: At 225M I get 1153GH/s using 1148W using an EVGA 1300W G2 on 115V. It will probably be a little better on 240V, but I don't have per-outlet metering on my PDU and I wanted to get power measurements using my KillAWatt.
To get it running @ +225Mhz I would suggest to up the voltage to 0.85V... Get out your pencil!! |
|
|
|
I think overclocking is going to be extremely limited, with four TPS53355 per board, or one for 16 chips. On the S1, it's one for 8 chips.
Thats the detail I've been waiting for  *really hoping for a PCB photo still* The S1 could consume around 45W per group of 8 chips overclocked (overclocking pushed the regulator right to its limits, but the real limiting factor was the voltage being set to 1.1V - some people have apparently pencil-modded the S1 to almost 215GH/~470W, or about 60w/8 chips) 45W/1.1V=40.9A on a 30A-rated component. (bitfury h-boards faced very similar amperage limits when overvolted around 0.95V) Its likely these chips run at around 0.8V and 1000W/640chips is about 25W per cluster of 16. That leaves a LOT of headroom on the regulators, but some of this is lost to the difference in voltage (more amperage). 25W/0.8V= 31.25A stock. Presumably these should be able to overclock without any issues from the regulator. I imagine all other components are beefy enough also. I imagine you will hit hardware errors at 15-20% overclock as with the S1. These may even be capable of up to 1.5TH is voltage was increased to ~0.9V 0.75V unloaded, so a little higher than that on the current. Still some room and I'd expect people should be able to get 20-40% overclocks, but not the 100+% ones we might have been able to do with 8 chips per vrm. I'm quite sure those puny heatsinks will not be enough when overclocked, 200Mhz+ will fry the chips unless some hefty cooling is applied! If you want to overclock the BM chip, the S1 was made just for that!! thats a losing attitude - anything can overclock! If it's running at 57 degrees @193MHz there's not much headroom!! |
|
|
|
I think overclocking is going to be extremely limited, with four TPS53355 per board, or one for 16 chips. On the S1, it's one for 8 chips.
Thats the detail I've been waiting for *really hoping for a PCB photo still* The S1 could consume around 45W per group of 8 chips overclocked (overclocking pushed the regulator right to its limits, but the real limiting factor was the voltage being set to 1.1V - some people have apparently pencil-modded the S1 to almost 215GH/~470W, or about 60w/8 chips) 45W/1.1V=40.9A on a 30A-rated component. (bitfury h-boards faced very similar amperage limits when overvolted around 0.95V) Its likely these chips run at around 0.8V and 1000W/640chips is about 25W per cluster of 16. That leaves a LOT of headroom on the regulators, but some of this is lost to the difference in voltage (more amperage). 25W/0.8V= 31.25A stock. Presumably these should be able to overclock without any issues from the regulator. I imagine all other components are beefy enough also. I imagine you will hit hardware errors at 15-20% overclock as with the S1. These may even be capable of up to 1.5TH is voltage was increased to ~0.9V 0.75V unloaded, so a little higher than that on the current. Still some room and I'd expect people should be able to get 20-40% overclocks, but not the 100+% ones we might have been able to do with 8 chips per vrm. I'm quite sure those puny heatsinks will not be enough when overclocked, 200Mhz+ will fry the chips unless some hefty cooling is applied! If you want to overclock the BM chip, the S1 was made just for that!! |
|
|
|
I don't get why a lot of people are complaining about the ethernet port placement. If you're going to rackmount it, isn't it easier to have ethernet in front just like many rackmount switches are setup?
Spondoolies-tech also has their ethernet port up front for their rackmount miners.
People trying to sound like they know anything about running a datacenter  I've seen ethernet and fiber setup various ways... all in front, front mount switch to back connected servers, and all in the back with backward facing switches. I kind of prefer the front mount connections if the wire routing is done "cleanly". Makes it easier to swap out servers... I guess it's just a personal preference. If you have to remove the unit from the rack, does it really matter? The powerplug is on the back and I guess you'll have to unplug that aswell... |
|
|
|
|
Depends on the temperature in the room, clockfrequency etc. But mine has never gone above 3000 RPM, even when running at 400MHz!
|
|
|
|
|
Sounds like your fan is at max RPM, could be a faulty fan and/or controller! But it will never be quiet as stated on the hp, it needs its own room!
|
|
|
|
|
