Topic: Avalon ASIC users thread - page 171. (Read 438724 times)

Hi,
I noticed that too and asked CKolivas. He replied:

The old avalon code lied about the hashrate and lied about the hardware errors. It counted hardware errors as hashrate (and since it's a hardware error it can't be valid hashes that it's doing) and didn't count "no matching work" scenarios as hardware errors. So the new code will appear to have a lower hashrate and a higher hw error count, but in fact it's doing more useful work and just not lying about the rates (along with all the other benefits in the new code).

https://bitcointalksearch.org/topic/m.2406402

I did. Accourding to my testing 3.2 clocked at 300 is making about 68000 compared to 3.1 70500. 24 hours run. I downgraded to cgminer 3.1. I do not know what is the reason, but there is performance decrease.

Will Anyone else to share the cgminer 3.2 results?

Anyone try the latest firmware?

Cheers

Can anyone report what is maximum safe operating temperatures for Avalon?

There are three temperature values reported Temp1, Temp2, Temp3. What do these correspond to?
With the current weather and temperatures I wanted to know at what cutoff point should I downclock the avalon.

Also, what is the max safe junction temperature of avalon chips?

so on circuit you can have 120 x 15= 1800 Watts maximum (not taking in account Cos phi effects and such)

Buy a kill-a-watt and mesure the exact consumption of your individual machines in action. Slap on some extra safety margin

For example, with this weather, my Batch #1 Avalon @ 300 MHz are consuming 655 Watts

So three machines would be > 1800 (and I would not try to load it to 99%)


I think you should also look at how practical it is to get rid of the excess heat, wherever you decide the put your machines


If you have enough circuits and enough room to vent, and you neighbours don't complain about the fan noise... yes it should be no problem

I can't seem to find the thread now, but recently someone posted a question regarding how many Avalons it is safe to plug into a typical outlet (I believe someone said two would be the max).  

I would like to set all this up in my house.  Here, we have 120 volt lines and 15 amp breakers.  This is the hardware I will have:

-three 3-module Avalons
-two 4-module Avalons
-two block erupter blades
-GPU farm (6 7950s) may also be running for a short while until they are not profitable.

I may end up distributing the load so I have each Avalon on a separate circuit, but even with that setup I want to make sure I am not approaching capacity of my panel.

Is all of this possible on a residential setup or do I need to look into leasing some space?

I've opened one of my two units the other day, batch 2, received may 31st and mining since then with wifi disabled, no signs of overheating and units stable and hashing well.

I've taken a couple of photos, just to be able to compare chip state later.

Flat cables from control unit to blades can block airflow though, so I bended them a little  to leave a passage for the air.

spiccioli

I removed the F1 fuses on my two Avalons this morning - It's surprisingly easy (DISCLAIMER - I'm not responsible for any issues), I just pinched off the fuses with my leatherman pliers.

All back up and running perfectly.

For anyone interested, after running for a week with the fuses, there was no noticeable damage to the chips on the control boards. However, it's still recommended to remove the fuses.

Fuse (F1) removal of AvalonV2 v1.5 controller board. USB hub chip runs considerably cooler with this mod.









http://imgur.com/a/88pys#0

The last one anti76 is a "special case" IIRC: he managed to get in batch1 by harassing Avalon with an order he didn't pay in time and I wouldn't be surprised (and wouldn't blame Avalon) if he would be the very last customer serviced at the end of batch3 (would probably make him wait less than an early customer of BFL ASICs has/will have)...

There are 6 others in the OP, which makes it 2+% of the 300 units to be delivered (assuming at least one missing Avalon per order and that these customers didn't forget to come back in the thread after receiving their units).
It's not really high for international shipping of something that customs can't make heads of tails (there were customer reports having been contacted by customs to explain what the package was and confirm how much it cost).

What's missing from the OP is the number of units, the problem isn't the same if we are talking of 10 units/order instead of 1. I'm waiting for one Avalon myself and would really like to know that.
Fortunately I seem to have a low-risk profile: I've used an account instead of doing an anonymous purchase (which seemed to make things easier for batch1), live in a country with not much corruption (at least from average people like the ones working in customs) and they had all my delivery information including my mobile phone number from the start (there were problems with missing delivery information and returned packages according to Avalon IIRC).

Investing in ASICs is clearly high risk (from various reasons), don't put all your coins in the same basket if you can...

suxx

Guys, why Avalon team is ignoring their users from 1st Avalon Batch, which haven't received their orders yet? https://bitcointalk.org/index.php?topic=157856.msg2356354;topicseen#msg2356354
I haven't got order here, but that seems to be pure wrong!

Here are the instructions previously in this thread

https://bitcointalksearch.org/topic/m.2132938

Google for openWRT and Failsafe.

Yes, wifi was and still is enabled and working.

Does anyone know how to rest the password on the Avalon unit? Each time I try to log in it says wrong password.


 

Yes, I think the Wi-Fi interface is working in the bridge mode in Avalon so it can still be powered while is 'not used'. And without antenna attached PHY layer have to ramp up its power to have more/less stable signal while just listening around. Xiangfu said that this happens not on all TP-LINK units so probably there is HW bug in some SoC revisions which is related to power management as well.

If anybody with fried USB hub chip care to share his antenna condition (attached or not) this would clear this guess up a lot.

WinTame2012:  Interesting idea about the wifi power draw being affected by the antenna.  I had installed 9dB antenna's on mine, which have now been decommissioned, of course.  Above, senseless says that he was not using wifi when the chip was burned, but it was still enabled.  No doubt he wasn't using an antenna, if he wasn't using wifi.  And that is the scenario during which his chip was burned.

spiccioli:  I think that under the 'interface' tab, where the LAN and WWAN devices are listed, it allows for disabling, if I remember correctly.  Under the wifi tab it offers an option to remove the instance that is setup, below the scan button.  I removed this instance and the WWAN listing on the 'interface' tab had then disappeared.  I am assuming the kernel doesn't load the wifi software/driver, if there is no listing there, but I don't know enough to check this.

ProfMac:  Yes, this is good news.  I do note that the decision to make this change had already been made by the time senseless was in the Avalon irc chat with xiangfu when he learned of the news of this.  I was there with senseless and abracadabra when he was being shown the pics of the burned chip and hearing senseless's news.  ngzhang had communicated to him already that F1 fuse removal was a good idea... and that the remaining batch2/3 unites were being made with the revision.

drakness

This is evidence that the team is tuning the design.
I myself am glad to know this.

Batch#2 was shipped w/o antenna...
I've catched a thought last night that chip overheating could take a place when TP-LINK have enabled Wi-Fi module but missing antenna. If TP-LINK implements dynamic power calibration for Wi-Fi PHY lacking of antenna could force it drawing much more power hence USB-hub chip overheating. This is just my deduction though.