|
MXRider
|
 |
May 23, 2012, 09:11:09 PM |
|
What software do can we expect to be using on these?
Everything that works with Icarus. CG Miner for example. |
|
|
|
|
|
Lethos
|
|
May 23, 2012, 09:14:41 PM |
|
What software do can we expect to be using on these?
Everything that works with Icarus. CG Miner for example. Happy with that, was expecting that to be an answer to come along. I'm sure that will be confirmed soon. |
|
|
|
|
punin
|
|
May 23, 2012, 10:02:26 PM |
|
Yohan, any estimate on when you will be able to post performance numbers?
|
|
|
|
|
sadpandatech
|
|
May 23, 2012, 11:01:35 PM |
|
It's the best looking fpga of the market, I cannot wait to get mine.
Lets hope performance reach 800Mh/s
Let's hope performance is 960MH/s minimum with Ed's bitstream or up to 1200MH/s+ with bitfury bitstreams. Hey, one can dream...... |
If you're not excited by the idea of being an early adopter 'now', then you should come back in three or four years and either tell us "Told you it'd never work!" or join what should, by then, be a much more stable and easier-to-use system.
- GA
It is being worked on by smart people. -DamienBlack
|
|
|
|
yohan (OP)
|
|
May 23, 2012, 11:05:11 PM |
|
We got first results today using one of the standard Icarus bitstreams and we do have a small problem using the standard bitstream in that we appear to have TX and RX crossed over incorrectly between the 2 FPGAs in the chain so we believe the second FPGA was not getting commands correctly. We will modify one board to check if this is correct over the next few days.
On the one half of pair that was working it was running at the 190MH/s that was expected of that particular bitstream using a standard clocking arrangement. We can sort out the TX/RX issue in a number of ways - a bitstream hack to swap TX/RX, a straight rebuild, or a whole new virgin bitstream option. We are working on all 3 of these options. It's more annoying than a major problem as it would have been nice for an Icarus bitstream just to drop in and work as a demo.
Todays test was purely about getting the bitstream and comms working and not in any way a performance test and it just ran at standard settings. We didn't do any serious measurements today on heat and power but the FPGA we were working with appeared to be in a good state and wasn't too hot.
There won't be any work on this tomorrow and probably a limited amount on Friday so don't expect much more until early next week.
Yohan
|
|
|
|
|
Garr255
Legendary
 Offline
Activity: 938
Merit: 1000
What's a GPU?
|
|
May 23, 2012, 11:10:01 PM |
|
That's better news than bad news. And: Do you have a pinout spec for all those little headers?
|
“First they ignore you, then they laugh at you, then they fight you, then you win.” -- Mahatma Gandhi
Average time between signing on to bitcointalk: Two weeks. Please don't expect responses any faster than that!
|
|
|
|
yohan (OP)
|
|
May 24, 2012, 04:35:22 AM |
|
Nothing formalised yet on the UP/DOWN but there will be and I will talk a bit more about planned functionality of this interface in a few days time. It will be a simple straight forward 1-1 wiring in the main up/down configuration but many other things are possible.
|
|
|
|
|
|
Lethos
|
|
May 24, 2012, 08:30:07 AM |
|
Good to hear some progress, Yohan. You said you used the Icarus bitstream, which is like firmware for these boards, right?
But what bitcoin mining software did you use? CGMiner?
|
|
|
|
Garr255
Legendary
Offline
Activity: 938
Merit: 1000
What's a GPU?
|
|
May 24, 2012, 06:35:16 PM |
|
Good to hear some progress, Yohan. You said you used the Icarus bitstream, which is like firmware for these boards, right?
But what bitcoin mining software did you use? CGMiner?
I know MPBM support Icarus, and I believe CG does as well. |
“First they ignore you, then they laugh at you, then they fight you, then you win.” -- Mahatma Gandhi
Average time between signing on to bitcointalk: Two weeks. Please don't expect responses any faster than that!
|
|
|
|
DILLIGAF
|
|
May 24, 2012, 06:42:14 PM |
|
I believe CG does as well.
It does just needs to be enabled when you compile it. |
|
|
|
|
|
bulanula
|
|
May 24, 2012, 08:48:33 PM
Last edit: May 25, 2012, 06:31:53 PM by Maged |
|
Why use bitpay and extra fees when you can convert your BTC to gbp and pay yohan with gbp by yourself manually ? Lazy or love automation !?
|
|
|
|
|
|
dirtycat
|
|
May 25, 2012, 01:59:27 AM |
|
 This is how the first 100 complete units will look. After then it 10mm lower. these are beautiful.. nice work! I would be proud to have a few of these sitting on my desk hashing away for the world to see. other fpga manufacturers need to take a look at these and learn a few things instead of trying to sell butt ugly overpriced junk EDIT: when they are able to start hashing  |
poop!
|
|
|
|
yohan (OP)
|
|
May 25, 2012, 11:58:51 AM |
|
That is exactly one our problems starting to use Bitcoin as a payment in a hurry. All of our bills are in USD or GBP so we would have to do that conversion. Then there is the admin on that multiplied by a couple of hundred customers.
Ok there wasn't much progress yesterday as expected although we nearly have our own first FPGA build yesterday. It has a few routes to complete and we might get to that later today. There will be limited progress today for various reasons but after that we will be back on full strength.
Fans, mounting pillars, stacking pillars, and pillar feet have now arrived in large numbers and we will show a picture of a pair of boards stacked, using the optional stacking pillars, hopefully later today. If we can make today's shipping cut-off a handful of boards will leave today to some of you who have requested developer status. If not these will go Monday.
The plan is currently that the main ship will start Monday or Tuesday. The first couple of days it will be relatively small numbers as we bed down the final assembly, testing, boxing and shipping processes. As with most projects this should gain momentum day on day probably for 2-3 weeks before we actually hit output and then things really start to shift.
The answer another question we are currently playing with CGMiner and that recognises the first chip in each of our pairs. The TX/RX issue is stopping the action of the second chip of the pair and we are awaiting our first build to complete so that can cease being an issue.
Proper benchmarks will need the second FPGAs up to be totally meaningful and that is coming but not for few days yet. Our main focus is to have a Cairnsmore1 boards loaded with a basic operational build. Higher performance bitstreams will come later.
Yohan
|
|
|
|
|
wildemagic
Member
Offline
Activity: 112
Merit: 10
|
|
May 25, 2012, 12:13:45 PM |
|
Higher performance bitstreams will come later.
Its good news that you plan to continue developing the bitstream for more performance in the future. kind regards |
.,-._|\ Offgrid 1.7kW Solar and 3G wireless internet powering my mining rig.
/ .Oz. \
\_,--.x/ [219.5btc of successful trades total] with : rastapool, miernik, flatronw & OneFixt
o
|
|
|
|
Lethos
|
|
May 25, 2012, 12:30:30 PM |
|
Good to hear an update Yohan on the software and hardware progress.
I've only just started using CGMiner, I've not noticed any performance difference to phoenix 2.0 on my current hardware (which I've used for a while), so it be nice to see a full set of quads working and the results which follow.
Thank you for keeping us updated regularly.
|
|
|
|
Turbor
Legendary
Offline
Activity: 1022
Merit: 1000
BitMinter
|
|
May 25, 2012, 04:35:44 PM |
|
 afraid your GPUs can't keep up anymore ? I remember your words about FPGA junk... |
|
|
|
rjk
Sr. Member
Offline
Activity: 448
Merit: 250
1ngldh
|
|
May 25, 2012, 04:38:28 PM |
|
afraid your GPUs can't keep up anymore ? I remember your words about FPGA junk... He's not actually sure of his position, as evidenced by endless fluctuation on the subject. But fence-sitters never get anywhere anyways. |
|
|
|
|
bulanula
|
|
May 25, 2012, 04:39:16 PM |
|
afraid your GPUs can't keep up anymore ? I remember your words about FPGA junk... He's not actually sure of his position, as evidenced by endless fluctuation on the subject. But fence-sitters never get anywhere anyways. I am damn sure about my position. I will be laughing in the end, when the gold dries up after the reward drop  |
|
|
|
|
|
norulezapply
|
|
May 25, 2012, 05:58:41 PM |
|
Great work keeping everyone so well informed yohan! Enterpoint is really doing an excellent job so far with Cairnsmore1 and showing other FPGA producers how it should be done professionally!
Even if these FPGAs offered less MHash/$£, I'd still buy them from Enterpoint over the competitors products any day, solely on the support you've given the community on the development process so far...
|
|
|
|
|
Turbor
Legendary
Offline
Activity: 1022
Merit: 1000
BitMinter
|
|
May 25, 2012, 06:06:07 PM |
|
Great work keeping everyone so well informed yohan! Enterpoint is really doing an excellent job so far with Cairnsmore1 and showing other FPGA producers how it should be done professionally!
Even if these FPGAs offered less MHash/$£, I'd still buy them from Enterpoint over the competitors products any day, solely on the support you've given the community on the development process so far...
Yohan does a great job ! To be fair, some competitors work alone or in a very small team. That leaves not much time for frequent updates. I own products of 3 different manufactors and all of them have been very helpful. |
|
|
|
|
