Bitmain Announces the Antminer R4 and APW5 Power Supply, Designed for Silence

[bit2255]

Great - Thanks!
BTW - How do I see what the current release and/or firmware I have? The system/overview info for both R4s appear to be identical....
Thanks again (sorry for the newbie questions)

System > Overview, look at the date after File System Version. That is actually the firmware version. For Bitmains downloads the date is given in YR-MO-DAY-24hr time format eg. 201612020035 is 2016 Dec 2 00:35 Eastern Time

So my 1st R4 is:Fri Dec 2 00:35:59 EST 2016

2nd R4 is:Sat Nov 19 03:13:24 EST 2016

Is the 1st considered an "upgrade" to the firmware? 

I read somewhere here NOT to fool with firmware upgrades..but the unit is running hot..hotter than a few hours ago..in the mid 116F ish right now

[1713557687]

1713557687

[1713557687]

1713557687

[1713557687]

1713557687

[1713557687]

1713557687

[1713557687]

1713557687

[NotFuzzyWarm]

Great - Thanks!
BTW - How do I see what the current release and/or firmware I have? The system/overview info for both R4s appear to be identical....
Thanks again (sorry for the newbie questions)

System > Overview, look at the date after File System Version. That is actually the firmware version. For Bitmains downloads the date is given in YR-MO-DAY-24hr time format eg. 201612020035 is 2016 Dec 2 00:35 Eastern Time

So my 1st R4 is:Fri Dec 2 00:35:59 EST 2016

2nd R4 is:Sat Nov 19 03:13:24 EST 2016

Is the 1st considered an "upgrade" to the firmware?  

I read somewhere here NOT to fool with firmware upgrades..but the unit is running hot..hotter than a few hours ago..in the mid 116F ish right now

Yes. As to *why* they are different and why your first one (as in bought first?) has newer firmware, who knows....

EDIT: Query -- which one runs hot? The Nov 19 or Dec 2 release? I ask because the R4 b6 that lost a hashboard (reports only 8 ASIC's) has the Dec 2 firmware and the remaining board runs around 100C.

The R4 b6 I got Monday has firmware dated Dec 9 and runs 3-5 C cooler.

[fanatic26]

There is no need to fear firmware upgrades as long as you confirm the firmware is for your device/batch/whatever before attempting to flash. I have flashed literally thousands of bitmain units without bricking a single controller.

[Longsnowsm]

Great - Thanks!
BTW - How do I see what the current release and/or firmware I have? The system/overview info for both R4s appear to be identical....
Thanks again (sorry for the newbie questions)

System > Overview, look at the date after File System Version. That is actually the firmware version. For Bitmains downloads the date is given in YR-MO-DAY-24hr time format eg. 201612020035 is 2016 Dec 2 00:35 Eastern Time

So my 1st R4 is:Fri Dec 2 00:35:59 EST 2016

2nd R4 is:Sat Nov 19 03:13:24 EST 2016

Is the 1st considered an "upgrade" to the firmware? 

I read somewhere here NOT to fool with firmware upgrades..but the unit is running hot..hotter than a few hours ago..in the mid 116F ish right now

As soon as you open the ticket with Bitmain they will tell you to update the firmware. It did correct the overheating situation I had on my R4.  Also others have reported it bringing back boards that disapeared.  So you really need to to this update.

[bit2255]

I am in the process of upgrading now.

One question - after I unzipped the R4 firmware from Antminer, there are 2 folders (c5 and xilinx) which contain 2 files, a bin file and the one i need to use (i believe) .. a file named "angstrom_rootfs.jffs2"

Both of these files are have the same date (12/2/2016) but have slightly different file-sizes.

The one in the c5 folder is 20,397 KB and the other one in the xilinx folder is 20,864 KB.

Which one do I use???

[HagssFIN]

DO NOT UNZIP THE ARCHIVE.
Upgrade it with the zipped file.
Please follow the Antminer R4 manual found at www.bitmain.com

[bit2255]

Upgraded of the firmware completed SUCCESSFULLY!

Both chis are being monitored, fan speed higher.

Temps are coming down, one still at 115F right now, other is down about 8 degrees so far..still low 100's but where it was.

No errors and hashing at capacity!

I'll keep a close eye on the temps ad let you know how it goes.

THANKS!!!!!!!

[NotFuzzyWarm]

Yep, right spot. Yours is dated Nov 25, 2016

Just as with my 1st batch-6 (with bad board after 22hrs so is 1/2 speed, Dec 2 firmware), this latest batch-6 has eerily smooth hashrate compared to the s9's.
ref:

I wonder if it has to do with this bit of code the log spits out and the miner is per-chip tweaking to give advertised spec?

Code:

miner total rate=8898GH/s fixed rate=8700GH/s

The R4 Batch-4 in my main farm at work is just as rugged looking as those s9's are. Hmm...

That graph looks more like the R4's are being logged on low difficulty shares and the S9's logging higher diff shares showing more variance. My R4's show similar spikes as my S7's & S9's.

No they are not. They are pointed at CKpool which holds a very steady diff over any given time vs some pools like Antpool where I'd see several k difference in difficulty when checking several times over say an hour.

The R4 batch-4 at work (often >9THs) shows diff of 7.078k and the 2x R4 b6 at home read 6.473k for the good one (rated 8.7THs, near dead-on that speed) and 3.377k on the one with 1 bad board. Bitmain's software is doing something else to even out the hash rates on a very fine-grained scale.

Even my s7's show more variance than these 2x R' batch-6's do.

[agentcash]

That's strange. I use kano (CKPool) as well and my worker shift graphs on the site show similar variance among all equipment I've looked at. If the software you're using is pulling the speed direct from the miner, that might explain it. The reported speeds in the antminer interface can vary quite a bit from actual pool-side speeds.

[VRobb]

Indeed, that eerily smooth graph data contradicts the variance that inevitably must be there...!  
Something fishy, thoughts?

[NotFuzzyWarm]

The graph is from Awesome Miner showing hashrate over 1hr. Highly recommend it!
Anywho the sampling rate is every minute and yes is read directly from the miner via the API so it is true throughput.

Now, as I recall, part of the 'Majik' in BTC software is that along with input from the pools the firmware/software in the miner itself sets the diff of work sent to each hashboard or ASIC. To me it looks like Bitmain is tweaking that internal diff on-the-fly more often and probably per-chip... Seems like a good way to throttle speeds. Given that they are using a top-end Xilinx processor the controller certainly has the horsepower to do it...

[tbonetony]

so far i am in possession of 7.5, 8, and 8.7 models. I noticed that batch 6 - 8.7 model runs significantly hotter than the other two. Specifically ranging 87 - 112 deg in readings in the same environment. I know it runs at higher Hz but this makes me concern its lifespan. anyone can echo on this?

[powrslave]

Just fired up first R4

Latest FW was on device
File System Version   Fri Dec 2 00:35:43 EST 2016

I was wondering if anybody had any thoughts or observations concerning this miner as I am a total newbie.


https://i.imgur.com/BuXtKOb.png

[philipma1957]

Just fired up first R4

Latest FW was on device
File System Version   Fri Dec 2 00:35:43 EST 2016

I was wondering if anybody had any thoughts or observations concerning this miner as I am a total newbie.

Your hard ware errors are a bit high on one board.  But not terrible.

Temps are decent fan speed is good.

I guess it is a 7.5 gh model.

Just keep an eye out on the hw 23 is okay but they can be 23 a week not 23 in an hour

[powrslave]

Your hard ware errors are a bit high on one board.  But not terrible.

Temps are decent fan speed is good.

I guess it is a 7.5 gh model.

Just keep an eye out on the hw 23 is okay but they can be 23 a week not 23 in an hour

Thanks! I guess I got a crappy one.

[NotFuzzyWarm]

Your hard ware errors are a bit high on one board.  But not terrible.
Temps are decent fan speed is good.
I guess it is a 7.5 gh model.
Just keep an eye out on the hw 23 is okay but they can be 23 a week not 23 in an hour

Thanks! I guess I got a crappy one.

Not necessarily.Quite often there will be a fair amount of HW errors within the 1st few hours but then hopefully things will settle down.

After 17.5 days of running my good R4b6 shows 326 HW errors -- and they are all on the 1st chain. 2nd board  -- 0 errors. On my bad one with a dead board, 356 errors in same time.

I've decided to screw the Warranty on it, I'll be doing some forensics on the dead board to maybe find wassup with it. Worse comes to worse it'll go out to BitmainWarranty in CO for Paid repair.

[IntroVert]

I think there are whole lot of different situations related to HW errors. The higher hashing board on my 8TH R4B3 always has 40-50 times of errors compared the other one. But the aggregate DiffA HW% is always < 0.001%.

At this very moment, system is up slightly more than 5 days, HW errors are 94 and 5167 and DiffA HW% 0.0006%. Temps are 8x/9x.

It seems a crapshoot on what kind of boards you get in your R4, as long they can be auto tuned to the miner's advertised hashrate. Not much consistency from what I see folks reporting on here.

[VRobb]

The DC where my malfunctioning droid R4B4 has been sputtering along at half speed has started the RMA process along with various S9s they have there too. Nice failure rate on these units, hopefully this doesn't take too long... 

[GWhisper]

The DC where my malfunctioning droid R4B4 has been sputtering along at half speed has started the RMA process along with various S9s they have there too. Nice failure rate on these units, hopefully this doesn't take too long... 

I sent my two R4's off last Tuesday, tracking indicates I should have them back tomorrow. I'm hoping for better luck this time around.

[powrslave]

Are there any APW5 PSU "clones" being made in the USA? Less expensive?

I know where I can buy low-cost 7 connector PSU's but I was looking for something similar to the APW5 PSU due to it's
large fan/design. very quiet compared to other psu's