 Show Posts
|
|
Pages: « 1 2 [3] 4 »
|
With current low prices for the coins an Acorn can help with, all the scenarios I've looked at show that they'll pay for themselves in 12-18 months. Is that what other people are coming up with too?
Can't wait to see some real $/day improvements when they ship: so many GPU/MB combo's to try!
Hmm they do x2 x3 the speed on some algorithms if your gpu's are set up right. If thats not real $ per day improvements I dont know what is hehe, the big problem right now is that we are in a major reversal trend. |
|
|
|
8/10 vs 128/130 has absolutely nothing to do with this. That is simply the encoding for DC Balancing/ensure there is a periodic bitflip on the transcievers so they can stay locked. It was changed in PCIe 3.0 due to improved transceivers and to lower overhead.
Also for everyone checking lspci - most GPUs now adays are designed for low-power idle. They run the link at x1 unless they are actually being used for transfer! Run a bandwidth heavy test or activity and run lspci, you’ll see LnkStatus kick up to 3.0 (8 Gbps)
We just spent weeks testing dozens of risers and slots on 3 dozen motherboards. I haven’t found a single riser that can’t sustain PCIe 3.0 speeds. While monitoring PCIe for any errors. I have, however, found crap “mining” motherboards that have PCIe errors on their slots even with a x1 device plugged straight in. I’ll take that actual testing over your anecdote.
Software (miner) and APIs: This is a Pre-Order for the hardware. The software will launch with the hardware shipping late August. We’ve already asked devs to reach out for API and SDK information that will be available before launch. Developer boards haven’t even been sent out yet to other devs.
I must have some real shitty risers then, because they refuse to do 3.0 in any of my boards including normal non mining boards, not only on the chipset's pcie ports, but also the cpu's x16 port  And for the lspci and idle thing, if you get 2,5 GT/s that could be it yeah, But im getting the 5 GT/s. And its while mining eth so the bus is used. And yes some chipsets have shitty pcie that will make errors with just a standard network card, and its not only on mining boards, ive seen it on normal b250 boards also. |
|
|
|
In the end its an xilinx ultrascale fpga with some boot chip that programs the fpga each time it starts up with an ip core for pcie communication and most likely other shit, and you got the jtag so you can just load your own bitstream on it doing Javascript acceleration or whatever  Something like the BCU1525, that comes with a x16 pcie interface can most likely do the same shit. Acorn is just a smaller version of it. The sqrl guys most likely got the idea from BCU1525, after they noticed it had a strong pcie interface, and applying some thinking outside the box. |
|
|
|
Yep, regular old USB risers with 24" blue USB cables. Here's a pic of this particular rig:
I'll go check out the actual version, but I believe they are either 006C or 007C ones.
The same on my B250 boards does nothing, gpu's still running 5 GT/s, so this could be good news for Z270 and perhaps H270 owners, the intel 300 series might do the same, maybe better pcie hub. |
|
|
|
root@miner011:~# dmesg | grep pcieport
[ 0.674260] pcieport 0000:00:1c.6: hash matches
This is quite intresting, is this just with the regular x1 risers what version, and how long is the usb3 cable ? |
|
|
|
|
Jep det er jeg..
Looks like you are extremely lucky and got the right combination motherboard and risers to do pcie 3.0, check this
dmesg | grep pcieport
It will report if there is any errors on your pcie bus, you could have errors that is corrected but once you meet one that cant be corrected it can crash.
It will look like this:
[556771.420395] pcieport 0000:00:1d.2: PCIe Bus Error: severity=Corrected, type=Physical Layer, id=00ea(Receiver ID)
[556771.420404] pcieport 0000:00:1d.2: device [8086:a29a] error status/mask=00000001/00002000
[556771.420410] pcieport 0000:00:1d.2: [ 0] Receiver Error (First)
[556801.961161] pcieport 0000:00:1d.2: AER: Corrected error received: id=00ea
|
|
|
|
I know on my win10 rigs I can use HWiNFO to see the speed. Is there a command line option under Ubuntu that will show me?
Prøv : lspci -vv | grep -E 'LnkCap' |
|
|
|
alll risers support 3.0 (x1). well if motherboard support only 2.0, than the riser will do 2.0 as well.
Holy shit that worked! Above screenshot is from Asus PRIME Z270-A. I have 6 rigs with those mobos, so I'll switch the other 5 to gen3. Ditto on the my 2 MSI Z270-A PRO mobos. Should also work on my 2 X99 based rigs I suppose. Once I get my Acorns, I'll drop them both in a 7x 1060 first, see how it does, then a 7x 1080Ti rig, to see which show the better increase in terms of $/day. I realize 2 Acrons will run at 100% all the time on a 7x 1080Ti rig, but it should lift them all by an equal amount, and just partially offload the GPUs, right? I would go check in your operating sysatem if they are really doing 3.0, because its highly unlikely that you get 3.0 with usb3 cable risers EDIT: In linux for example this command would show your negotiated pcie speeds. lspci -vv | grep -E 'LnkCap' |
|
|
|
alll risers support 3.0 (x1). well if motherboard support only 2.0, than the riser will do 2.0 as well.
Its simple the usb3 cables dont have the quality to negotiate pcie 3.0, think about it usb3 cables are made for 8b/10b same as what pcie 2.0 run, but pcie 3.0 use 128b/130b. Now there is faster standards like USB 3.1 Gen2 or USB 3.2 that do use 128b/130b but i would imagine that they also have much shorter cables, and its not cheap  EDIT: Already at pcie 2.0 these usb risers suck, if you have problems with gpu fallen of the bus errors in your syslog, and your gpu's are not too much clocked, you can usualy fix it by forcing pcie 1.0, so the risers are iffy even for 2.0, you say they can 3.0 I say NO ! |
|
|
|
So some risers support pcie 3.0? Most of my recent risers are VER 008C, so maybe they do. When I initially setup my rigs, the guides all said to configure for pcie 1.0 or 2.0. I'll change the 7x 1060 rig with the Z270 mobo to gen3 and see what happens.
Also, I'm curious if the sqrl miner is free. I know it doesn't have a dev fee, which is nice.
Yes there is risers with pcie 3.0 but they use ribbon cables and is quite short, its not the usb3 cables you know from your regular pcie 2.0 risers. Im only guessing here, but if sqrl miner is not free, I would asume that having a acorn would also give you a license for the miner. |
|
|
|
|
1:2 would most likely be fine for your 1060 and 1070 rigs, and then you can use your x1 ports but you need pcie 3.0 risers, and a motherboard with intel 200+ chipset to get pcie 3.0 on the chipset.
But for your TI rigs you most likely need 1 acorn per gpu, so each gpu needs atleast a x2 3.0 or x4 2.0 pcie.
|
|
|
|
"but anyways if it's a 2.0 riser max 4gpu to 1 acorn density, if it's 3.0 link speed then it's up to 2 gpu to 1 acorn density. But the proper balance for a particular algo means doing the math with the numbers on the chart
if you are on a full x4 connection then you can go up to 1:1 density
or more if the gpu on x8 or w/e
1:2"
Sounds like the miner software will itself configure it to the best it can with the speeds you give it, thats good news, When its not stated on the site you think that a misconfiguration would result in a total bottleneck slowing down your hashrate to posible under what you would get without the acorns |
|
|
|
There's no way the LGA1151 socket has enough pcie lanes coming from the chipset to run that many GPUs on the 3.0 protocol.
Ehm, the x1 ports those mining boards have is from the chipset. the cpu just run the big x16 port. So in my case it would be 4x acorns on pcie 3.0 x16 via the cpu's pcie hub on the x16 port. And then data over dmi to the x1 ports to the gpu's. EDIT B250 have 12 pcie 3.0 lanes. |
|
|
|
|
There is one thing I keep thinking, the Acorn Nest X4 can run from a x16 2.0 interface, so if each acorn fpga card can boost 2 lower end gpu's like 1060 1070 and RX570 you should be able to use for example TB250-BTC PRO its x1 ports runs pcie 3.0, so 8 gpu in 8 of those x1 3.0 ports fits the speed the acorns is gona sit on with its x16 pcie 2.0 limit.
|
|
|
|
|
Looks like the cheapest solution for a big setup will be the "Nest X2G"
You could have 3x Nest X2G 6x acorn and 6x gpu, and it will fit fine in any motherboard with 3x pcie x4 3.0.
|
|
|
|
The info is just to little on these devices, like for example it does not tell if you can use the 2 bigger ones for ethereum or moneo, with only 50% more in price i guess more than me would be intrested in buying the bigger ones, using them for eth, for then to be able to fall back to something else when eth goes pos. It does not tell the exact bandwidth requirements so you can properly build a system it wont bottleneck in, the only thing we really know is that 4 pcie 2.0 lanes should be enough, and we can then only guess from what people here say for example that it "should" be able to boost 2 1060's, and if so then x2 slots must be required for them. Its a shame but I think im gona skip this one and check back in in august. |
|
|
|
Why not set up 32 Acorn CLE 215+s on a octominer and let them mine by themselves using little power.
I think most are looking to boost their gpu's for a little reinvestment, the solo mining is a bit iffy since pretty much anything an fpga can solo mine, you can make an asic for, so its highly speculative |
|
|
|
|
Hmm I did some digging and did some numbers, its a shame there is no performance numbers out on ethereum yet.
1, Acorn can run from at full speed on x4 pcie 2.0 slot without any limits.
2, Technicaly you should be able to use something like the AsRock ULTRA QUAD M.2 CARD in a x16 slot to feed 4 acorns. But another sulution with its own pcie switch would most likely be better supported since not all motherboards can split the x16 slot into 4 x4 lanes.
3, So 4 acorns should be able to boost without bottlenecks 8 gpu's each on x2 pcie 2.0 slots.
Whats left todo.
4, Specs says the 2 bigger ones is not suited for ethash, is that just because they are OP for that, can you buy the bigger ones mine ethash now, and have that extra security when ethash goes pos?
5, In case of such a setup on most cheap boards the acorns would be on the processors pcie hub, and the gpu's on the chipsets pcie hub, they link together via dmi as faar as i remember, can it even handle these speeds?
6, Find some cheap motherboard with x8 slots.
Or am i on the complete wrong track here ?
|
|
|
|
ETH would be a smart choice, especially if you are not forced to periodically sell your mining earnings, and can hold on till the price recovers a bit. These current prices
can be considered a solid bottom for robust coins like ETH & ZEC.
Exactly, and in the end you also have to say to yourself how do I want things to be, I dont want 10000 shitcoins being arround, now i dont call XMR a shitcoin, but manny of the other coins some waste their hashing power is shitcoins, and will most likely never be worth much so they are speculative mining something thats most likely never going up unless the big boss bitcoin pulls it up. As long as some support mine buy etc shitcoins they are gona be arround. What they should be doing was to mine one of the big established coins, and show their support to something that has proven itself. Its all about the greed, people think they came to late for ETH "now i better mine this shitcoin and hope it goes 1000x", well let me be clear here it most likely wont, and you would be better of mining ETH and buying your shitcoin since eth is what gives the most for most common graphics cards. |
|
|
|
because ~95% of the miners are retarded and dont know what they are doing
yes everybody who mines directly eth is either stupid or have free electricity, if your 580´s are not good on cryptonight your biosmod and/or mienr config sucks,
buy a fucking wattmeter then you will see the difference between ethash and cN, 850mv-875mv says a fuck about real powerdraw on different algos.
you need alot to learn
just a small example with my 4 vega rig
with proper tuning/config
4 vegas = ~180mhs eth = 5,59$ daily befory electricity(~950w from the wall whole rig)
4 vegas = 8000hs xmr = 6,49$ daily before electricity(~710w from the wall whole rig)
(profits i took current stats from whattomine for idiots)
cn = -25% powerdraw + 16% more $ daily on xmr + 25% saved electricity
Heh you dont have to be rude man, and XMR is only that good for cards with HBM. And you can say what you want, but in my opinion those most stupid are those with vega's, at the time they were released you should have known the bubble would burst So dont try to tell others they are stupid by showing your own stupidity. |
|
|
|
|
