ANTMINER S7 is available at bitmaintech.com with 4.86TH/s, 0.25J/GH - page 114.

philipma1957

legendary

Activity: 4256

Merit: 8551

'The right to privacy matters'

Quote from: VirosaGITS on January 01, 2016, 02:58:32 PM

Quote from: SLEI on January 01, 2016, 02:54:20 PM

Quote from: Prelude on January 01, 2016, 12:35:42 PM

Quote from: SLEI on January 01, 2016, 11:47:11 AM

Have anybody experience of zwilla.de firmwares for S7 they have very big promises?

Never heard of them. What are they claiming their firmware does?

20% more power (GHS?) 30% less power usage.

Firmware that magically improve efficiency by 56%(0.16J/GH)? Yeah i don't think so.

Well the batch eight went into the other direction about .3 watts vs .25 watts.

Based on some testing I did with early batch I am pretty sure the freq 575 batch 2 could run at .2 watts at freq 550 .

My grip with the s-7 is the batches vary so fucking much that you don't know want you are buying until you get your hands on it.

@ richbc I think the 200 watt difference he is talking about is on the same batch number order a few days apart.

I had this happen with batch 1 I had a dog at 4500 and a good one at 4800.

@ bitmaintech no comp no coupon really a lack of honor on your part.

Sad to see you act like that.

RichBC

hero member

Activity: 588

Merit: 500

Quote from: SLEI on January 01, 2016, 02:50:47 PM

Quote from: RichBC on January 01, 2016, 02:26:13 PM

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

Batch 6,8,9 are all 135 chip. Efficiency should be similar so long as the Buck Converter is set to the same voltage, which it may well not be, in all batches?

Rich

Miners can't be identical if batch 8 can have 200W difference in power usage at same frequence.

Which other Batch are you comparing the Batch 8 Miner to?

Rich

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: SLEI on January 01, 2016, 02:54:20 PM

Quote from: Prelude on January 01, 2016, 12:35:42 PM

Quote from: SLEI on January 01, 2016, 11:47:11 AM

Have anybody experience of zwilla.de firmwares for S7 they have very big promises?

Never heard of them. What are they claiming their firmware does?

20% more power (GHS?) 30% less power usage.

Firmware that magically improve efficiency by 56%(0.16J/GH)? Yeah i don't think so.

marvykkio

hero member

Activity: 798

Merit: 1000

to me they should send 3X S7 day 4, we see if it is true

SLEI

member

Activity: 84

Merit: 10

Quote from: Prelude on January 01, 2016, 12:35:42 PM

Quote from: SLEI on January 01, 2016, 11:47:11 AM

Have anybody experience of zwilla.de firmwares for S7 they have very big promises?

Never heard of them. What are they claiming their firmware does?

20% more power (GHS?) 30% less power usage.

SLEI

member

Activity: 84

Merit: 10

Quote from: RichBC on January 01, 2016, 02:26:13 PM

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

Batch 6,8,9 are all 135 chip. Efficiency should be similar so long as the Buck Converter is set to the same voltage, which it may well not be, in all batches?

Rich

Miners can't be identical if batch 8 can have 200W difference in power usage at same frequence.

dontetris

newbie

Activity: 47

Merit: 0

Did anyone receive the "late shipping compensation" yet?

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: RichBC on January 01, 2016, 02:37:02 PM

Quote from: VirosaGITS on January 01, 2016, 02:30:01 PM

Can a buck converter provide variable voltage depending on the clock the miner is set at? Some people said by underclocking it raised their efficiency and that should not happen unless the voltage is variable to an extent or there is more efficiency loss at higher speed.

If it is tied into the main controller then it could be adjusting the voltage, until someone traces the circuit / makes some measurements we will not know.

But agreed as we have all discovered just changing the clock frequency and leaving the voltage the same leaves the efficiency virtually the same. What sort of improvement to efficiency have people seen?

Rich

I look forward seeing whether or not the people who said you could underclock to increase efficiency were right. I did not say anything then because i'm not sure its untrue. Maybe they think raising efficiency is just lowering hashrate/clock/consumption.

But we'll see, i think phil or whoever gets a B8 to do test before me will do it, if not i will test it when i receive mine in a week or two(?).

RichBC

hero member

Activity: 588

Merit: 500

Quote from: VirosaGITS on January 01, 2016, 02:30:01 PM

Can a buck converter provide variable voltage depending on the clock the miner is set at? Some people said by underclocking it raised their efficiency and that should not happen unless the voltage is variable to an extent or there is more efficiency loss at higher speed.

If it is tied into the main controller then it could be adjusting the voltage, until someone traces the circuit / makes some measurements we will not know.

But agreed as we have all discovered just changing the clock frequency and leaving the voltage the same leaves the efficiency virtually the same. What sort of improvement to efficiency have people seen?

Rich

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: RichBC on January 01, 2016, 02:26:13 PM

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

Batch 6,8,9 are all 135 chip. Efficiency should be similar so long as the Buck Converter is set to the same voltage, which it may well not be, in all batches?

Rich

Can a buck converter provide variable voltage depending on the clock the miner is set at? Some people said by underclocking it raised their efficiency and that should not happen unless the voltage is variable to an extent or there is more efficiency loss at higher speed.

RichBC

hero member

Activity: 588

Merit: 500

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

Batch 6,8,9 are all 135 chip. Efficiency should be similar so long as the Buck Converter is set to the same voltage, which it may well not be, in all batches?

Rich

Seba13xp

newbie

Activity: 27

Merit: 0

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Quote from: Biodom on December 29, 2015, 02:10:11 AM

Quote from: dmwardjr on December 29, 2015, 01:45:42 AM

Quote from: Prelude on December 29, 2015, 01:21:23 AM

At 700MHz they're pretty much tapped out, and overclocking is not an option until we figure out how to control voltage. My OG S7s run at 6TH/s in comparison with overclocking.

I totally agree with you. I'm coming close to making the decision to go ahead and under clock these back to 600 MHz from the 700 MHz factory setting. If I'm not mistaken, the batch 6 (135 chips) were clocked at 600 MHz and hashed at 4.0 TH/s? I'm thinking about treating the new 135 chip rigs as batch 6 and under clock all of them to 600 MHz for longevity.

I got batch 6...it is totally stable being overclocked to 625, a bit less stable (or maybe less efficient?) at 650, so I will try either of these two instead of just 600, if you have room on PSU. The funny thing, if you remember as everyone was saying that B1 "likes' heat-~60C. In my observation, B6 "likes" being cold-I got better speed and less errors at 41-45C (same goes for B7, interestingly).

625-4.15-4.2 TH (1090W at the wall); 4.3 TH at 650 (1150W)

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

S7 Batch 6 Specifications:

1. Hash Rate: 4.05 TH/s

2. Power Consumption: 1042 W + 10% (at the wall, with APW3, 93% efficiency, 25C ambient temp)

3. Power Efficiency: 0.25 J/GH + 10% (at the wall, with APW3, 93% efficiency, 25°C ambient temp)

4. Rated Voltage: 11.60 ~ 13.00V

5. Chip quantity per unit: 135x BM1385

6. Dimensions: 301mm(L)*123mm(W)*155mm(H)

7. Cooling: 2x 12038 fan

8. Operating Temperature: 0 °C to 40 °C

9. Network Connection: Ethernet

10. Default Frequency: 600M

nhando

full member

Activity: 196

Merit: 100

Quote from: dmwardjr on January 01, 2016, 02:10:27 PM

Quote from: Biodom on December 29, 2015, 02:10:11 AM

Quote from: dmwardjr on December 29, 2015, 01:45:42 AM

Quote from: Prelude on December 29, 2015, 01:21:23 AM

At 700MHz they're pretty much tapped out, and overclocking is not an option until we figure out how to control voltage. My OG S7s run at 6TH/s in comparison with overclocking.

I totally agree with you. I'm coming close to making the decision to go ahead and under clock these back to 600 MHz from the 700 MHz factory setting. If I'm not mistaken, the batch 6 (135 chips) were clocked at 600 MHz and hashed at 4.0 TH/s? I'm thinking about treating the new 135 chip rigs as batch 6 and under clock all of them to 600 MHz for longevity.

I got batch 6...it is totally stable being overclocked to 625, a bit less stable (or maybe less efficient?) at 650, so I will try either of these two instead of just 600, if you have room on PSU. The funny thing, if you remember as everyone was saying that B1 "likes' heat-~60C. In my observation, B6 "likes" being cold-I got better speed and less errors at 41-45C (same goes for B7, interestingly).

625-4.15-4.2 TH (1090W at the wall); 4.3 TH at 650 (1150W)

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

135 Chips. See website info below

5. Chip quantity per unit: 135x BM1385

https://bitmaintech.com/productDetail.htm?pid=00020151105082702881F8f0B9I405FF

Seba13xp

newbie

Activity: 27

Merit: 0

Quote from: VirosaGITS on January 01, 2016, 02:09:16 PM

Quote from: Seba13xp on January 01, 2016, 02:02:40 PM

Was there ever a fw that allowed to hook up 9 blades to a single controller? I know that philipma1957 has 6 and that the new fw dont allow more than 3.

Technically, the S5+ controller allowed 9 boards but depending on which firmware you had on it, you would have 3 of the 9 boards "ghost boards". Not showing up on the WebUI.

That is why Im asking which fw allowed all 9 to show up and hash.

dmwardjr

legendary

Activity: 1302

Merit: 1318

Technical Analyst/Trader

Quote from: Biodom on December 29, 2015, 02:10:11 AM

Quote from: dmwardjr on December 29, 2015, 01:45:42 AM

Quote from: Prelude on December 29, 2015, 01:21:23 AM

At 700MHz they're pretty much tapped out, and overclocking is not an option until we figure out how to control voltage. My OG S7s run at 6TH/s in comparison with overclocking.

I totally agree with you. I'm coming close to making the decision to go ahead and under clock these back to 600 MHz from the 700 MHz factory setting. If I'm not mistaken, the batch 6 (135 chips) were clocked at 600 MHz and hashed at 4.0 TH/s? I'm thinking about treating the new 135 chip rigs as batch 6 and under clock all of them to 600 MHz for longevity.

I got batch 6...it is totally stable being overclocked to 625, a bit less stable (or maybe less efficient?) at 650, so I will try either of these two instead of just 600, if you have room on PSU. The funny thing, if you remember as everyone was saying that B1 "likes' heat-~60C. In my observation, B6 "likes" being cold-I got better speed and less errors at 41-45C (same goes for B7, interestingly).

625-4.15-4.2 TH (1090W at the wall); 4.3 TH at 650 (1150W)

Biodom or Anyone else,

Help me remember... Did Batch 6 have 135 chips or 162 chips? My question was never really answered about number of chips. I'm curious if I would get similar efficiency with under clocking as Biodom mentions here with Batch 6 but similar with Batch 8?

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: Seba13xp on January 01, 2016, 02:02:40 PM

Was there ever a fw that allowed to hook up 9 blades to a single controller? I know that philipma1957 has 6 and that the new fw dont allow more than 3.

Technically, the S5+ controller allowed 9 boards but depending on which firmware you had on it, you would have 3 of the 9 boards "ghost boards". Not showing up on the WebUI.

Seba13xp

newbie

Activity: 27

Merit: 0

Was there ever a fw that allowed to hook up 9 blades to a single controller? I know that philipma1957 has 6 and that the new fw dont allow more than 3.

philipma1957

legendary

Activity: 4256

Merit: 8551

'The right to privacy matters'

Quote from: lightfoot on January 01, 2016, 12:36:58 PM

Hm. Are s7's worth my time to learn to fix for other people?

Most of the miner problems seem to be the same things: Heat and power distribution. Thoughts?

your skill set. your tool set. if both are good yeah I would say go for it.

lightfoot

legendary

Activity: 3164

Merit: 2258

I fix broken miners. And make holes in teeth :-)

Hm. Are s7's worth my time to learn to fix for other people?

Most of the miner problems seem to be the same things: Heat and power distribution. Thoughts?

Prelude

legendary

Activity: 1596

Merit: 1000

Quote from: SLEI on January 01, 2016, 11:47:11 AM

Have anybody experience of zwilla.de firmwares for S7 they have very big promises?

Never heard of them. What are they claiming their firmware does?

Topic: ANTMINER S7 is available at bitmaintech.com with 4.86TH/s, 0.25J/GH - page 114. (Read 528055 times)