Bitcoin Forum
November 13, 2024, 10:45:12 AM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: 4 Modules Avalon Miner Open-box Report [Manufactured by Rain team from China]  (Read 1635 times)
windleaf (OP)
Newbie
*
Offline Offline

Activity: 11
Merit: 0


View Profile
August 15, 2013, 06:56:57 AM
 #1

The day I received phone call from SF Express guy, which made me a little exciting. When the machine delivered, the express guy asked me: "Is the sender your relative? Is that why he send 25kg package which made me so tired to you in Valentine's day?" (13th Aug is Chinese Valentine's day).
http://h.hiphotos.bdimg.com/album/s%3D900%3Br%3D90%3Bq%3D90/sign=598162d9b7003af349bad0600511b461/0bd162d9f2d3572c7e8d1e0c8b13632763d0c3f7.jpg

To open the wood frame is a tough work, which took me 40 minutes and hands full of hurt:
http://d.hiphotos.bdimg.com/album/s%3D1200%3Br%3D180%3Bq%3D90/sign=51e041e3728da9774a2f82298061c36b/8694a4c27d1ed21b1584ac35ac6eddc450da3f94.jpg

Here is the paper box:
http://h.hiphotos.bdimg.com/album/s%3D900%3Br%3D90%3Bq%3D90/sign=73a726c396dda144de0960b2828ca29f/3c6d55fbb2fb43161252f8aa21a4462308f7d395.jpg

Fillers for anti-shock are sufficient:
http://e.hiphotos.bdimg.com/album/s%3D900%3Br%3D90%3Bq%3D90/sign=32f863df8694a4c20e23eb2b3ecf69e8/9825bc315c6034a8baaf3a51ca13495408237695.jpg

The miner is carefully packed with air bubble film, whole set included machine and power line:
http://h.hiphotos.bdimg.com/album/s%3D900%3Br%3D90%3Bq%3D90/sign=6e988e3bb7fd5266a32b30149b23e516/9358d109b3de9c823e0b7b5f6d81800a18d843f0.jpg

Linked with power and internet to test if it was running:
http://h.hiphotos.bdimg.com/album/s%3D1200%3Bq%3D90/sign=1bd537ea18d8bc3ec20802c8b2bb9d6f/63d0f703918fa0ec7faee6dd279759ee3c6ddbf1.jpg

Configuration finished, the miner was mining with fans' roar. So I put it under desk immediately:
http://b.hiphotos.bdimg.com/album/s%3D900%3Br%3D90%3Bq%3D90/sign=2272d530f603918fd3d131ca610654aa/79f0f736afc379314b0004f9eac4b74542a911f1.jpg

Here comes the highlight, the secret gift from Rain team is......DUREX! In Valentine's day, oh my~~~;D (Wanna know what's said on the note? Ask your Chinese friends...):
http://d.hiphotos.bdimg.com/album/s%3D1200%3Bq%3D90/sign=c28f60ca6159252da717190604ab384a/8694a4c27d1ed21b1582ac35ac6eddc450da3f92.jpg

Now the miner was running over 24 hours, here's performance report. Below is default cgminer config:
http://d.hiphotos.bdimg.com/album/s%3D1600%3Bq%3D90/sign=daea3fb14ec2d562f608d4ebd721ab9e/f703738da9773912d0c914b0f9198618377ae270.jpg

Choose over clocking 350,the hash speed is aroud 103 or 104 GH/s in btcguild pool, the temperature and fan speed is stable:
http://c.hiphotos.bdimg.com/album/s%3D1600%3Bq%3D90/sign=ddda04a5ac345982c18ae1943cc40adc/d01373f082025aafee08ee1afaedab64034f1a2c.jpg

Here is a comparison, worker 1 is 4*USB block erupters, worker 2 is Avalon 4 modules miner ( Pool end speed is around 102 GH/s):
http://c.hiphotos.bdimg.com/album/s%3D1600%3Bq%3D90/sign=e259d30d55e736d15c13880eab6074b3/cefc1e178a82b90110c75ee2728da9773812effd.jpg

Btcguild provides graphic chart, the 2 works' speed gap is huge, work 1 is crawling on the ground:
http://b.hiphotos.bdimg.com/album/s%3D1600%3Bq%3D90/sign=27882550ca1349547a1eec62667ea929/6f061d950a7b020802a3a23063d9f2d3562cc8fd.jpg

In the end, let's have a look at profit. Recently the calculation power increased insanely, which decreased the profitablity per GH/s a lot. However, don't be sad if you believe in BTC's future:
http://c.hiphotos.bdimg.com/album/s%3D1600%3Bq%3D90/sign=026675a2ac4bd11300cdb3346a9f9f7e/a044ad345982b2b71a51a81530adcbef77099b11.jpg

Ok, here's the end of 4 Modules Avalon Miner Open-box Report. Wish Rain team brings us more and more better products. Wish you guys all have fun with BTC!

by Windleaf 15th Aug, 2013
mjeva
Newbie
*
Offline Offline

Activity: 46
Merit: 0


View Profile
August 15, 2013, 07:54:48 AM
 #2

Well,high price chips.
dethdeks
Full Member
***
Offline Offline

Activity: 206
Merit: 100



View Profile
August 15, 2013, 08:07:25 AM
 #3

why so many rejects on miner 2? kinda seems like something may be wrong. miner 1 has an acceptable amount of rejects but almost 20k for just one miner is something id be looking into.

Donations Accepted
LTC: Lcv7XmxnHJ7BhRjXjzCUBJ4Cfrq43SA5qc
BTC: 16mHGGVgntqtTmrddud8BFgN3XAsBobvVe
windleaf (OP)
Newbie
*
Offline Offline

Activity: 11
Merit: 0


View Profile
August 15, 2013, 08:38:05 AM
 #4

why so many rejects on miner 2? kinda seems like something may be wrong. miner 1 has an acceptable amount of rejects but almost 20k for just one miner is something id be looking into.

Maybe try reboot?
cryptograd
Member
**
Offline Offline

Activity: 112
Merit: 10



View Profile
August 15, 2013, 03:10:51 PM
 #5

why so many rejects on miner 2? kinda seems like something may be wrong. miner 1 has an acceptable amount of rejects but almost 20k for just one miner is something id be looking into.

Because the machine is over clocked

Liked something I said ->17ry6rrknqmQ2S1NRArzdrNMmG2Zk449AE
Most important bitcointalk post in history
https://bitcointalk.org/index.php?topic=120184.msg1381739#msg1381739
windleaf (OP)
Newbie
*
Offline Offline

Activity: 11
Merit: 0


View Profile
August 15, 2013, 03:43:15 PM
 #6

why so many rejects on miner 2? kinda seems like something may be wrong. miner 1 has an acceptable amount of rejects but almost 20k for just one miner is something id be looking into.

Because the machine is over clocked

ya, i think so. reboot didn't make improvement.
eleuthria
Legendary
*
Offline Offline

Activity: 1750
Merit: 1007



View Profile
August 15, 2013, 06:37:02 PM
 #7

Avalons firmware produces more stale shares than alternative miners (Block Erupters, FPGAs, GPUs running at moderate intensity), similar to how the BFL FPGAs (not sure on ASICs) did.  This is part of the reason they were terrible at working with p2pool (although some improvements to p2pool have been made to apply a moderate fix).

My understanding is the reason for this is the strings of chips are not asynchronous.  There is no way to interrupt their current job, so when a longpoll/new block message hits, the chips will finish their current piece of work before moving to the new one.  As a result, they'll be submitting stales from that previous work since there was no way to force them to update mid-process.

RIP BTC Guild, April 2011 - June 2015
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!