Topic: Acorn M.2 FPGA based GPU Accelerator - page 39. (Read 73507 times)

Fair enough, can we get hard numbers on a few algorithms for the 215+ now stand alone. If we are facing delays some info to bide the time would be at least informative.

Discord link https://discord.gg/QzQRd3

whats the discord channel?

It was always advertised as a pre-order, shipping late August.  I understand the sentiment - we certainly had hoped late August wouldn’t be quite the very end, but to be honest with the number of issues and supply chain glitches that occurred along the way I am very happy with where we are with progress.

We are very transparent and straight forward as a company - but the updates don’t always get posted here as often as they do in the discord, partially because I don’t prefer arguing with people.

The price for an Artix 7 FPGA with M2 interface is not bad   .
I hope we can buy it soon in the EU.

I completely agree. Aug 31 is just playing number game. No updates on the CLE 215+ solo mining. No information about SQRL miner a week before the release. Without proper mining pools support (including NiceHash strata) this whole project is not worth anything.

This is nice to see, not to be a politically correct asshole though but August 31st was never mentioned to us august buyers. It was "end of august" so this was really leveraging the timeline in literal terms. It also more than likely makes september preorders even more lucrative while shafting your first customers on mining time.

Just some thoughts from an early buyer of these. When delays are known post it immediately so we arent wondering why its end of week one September and our acorns are still not delivered.

 I hope the standalone mining of the cle 215+ blows my mind. It was advertised as being stand alone mining vs. The lower spec cheaper acorns , so when i see people ask and the responses the squirrel team gives on its solo mining capability they were less than comforting. Which again was another catch for us early buyers. So i really pray this was a case where you guys are just under promising so you can over deliver. If not then expect major disappointment from 215+ buyers.

If the acorns from august are delivered in September then they are late, buyers bought August expecting August.

Questions:

1a) Will either Nest board (2X or 4X) still contribute acceleration if not fully populated with Acorns?

1b) In essence, is the Acorn on a Nest board treated as a shared resource among the GPU's plugged into the USB socket or are they tied to a specific GPU?

1c) Does the Nest board (with Acorns) accelerate GPU's tied to the PCIe buss?

TIA

There's an update on Acorns:



This is the first Acorn CLE101 off the PCB production line.
Quote from GPUHoarder a few minutes ago

Living in UK i was initially shocked by the minimum shipping prices, when i ordered the CLE-215+ in late July.

My initial thoughts being they are using shipping as additional profits.

But on reflection, I'd rather see this start up company suceed in their endeavours and provide the community with further products and support.

People don't work for free.

Good luck guys. and can't wait for my product in september.

The pricing over at squirrelsresearch is indeed completely absurd for shipping. Right now, i'm still waiting for the solo performances to think about buying a few but in any case I will purchase via a Resend company because of this.

Hi all, So basically, Not sure what to buy. have abuild of 11x RX 560 2gb CN normal (i never mine core intense algo)
Another rig is Cryptonight heavy with Rx 580 8gb High end cards (8 cards)....
I'm intersted in CN normal/Heavy and Maybe Ethereum...
How many CLE101 are sufficient for most of the bonuses to apply.
Is the 215 (not the + variant) worth it, the double memory?
So Basically those cards aren't Card based but more of IO (hash amount based?) so that means a single CLE 101 can boost 4x 560 2gb or 2x 570s/580s ... What about CN heavy, thats massive bandwidth here to make 1150 h/s with an 8gb polaris (when compared to VEGAs tiny 1400 h/s in heavy)

Appreciaste any input.
Also, whats wrong with the shipping? One CLE101 cost 75USD to reach Australia (Business class 50 gram item ?) my concerns are all about the feasibility of mining the cost of these things are what you can pay right now for 3-5 Ethereum!

Any more info about standalone mining algos/estimated hashrate? I'm thinking about selling most/all of my GPUs so I'm hoping I'll be able to mine with these still.

Here we go again, please read https://www.reddit.com/r/Monero/comments/8b8o05/cryptonight_v7_vs_v1_a_nitpick_request_to_use_the/

Hope someone can explain please..... These do not boost CN v7 (heavy/light etc.) cause of its anti asic modifications? provided that i have an m2 MOBO in the first place, the acorn will support how many GPUs (full 13  AMD?)
Those FPGAs or asics or mining machines, how are they GPU friendly? seeing that the fans/developers are some of those promoting GPU mining (ohgodapeople) These are no accusations or any thing, I'm simply mining CN v7/heavyand atm ridiculously confused, trying to buy some bargain 8gb GPUs but meanwhile, seeing this mainly the -60% power saving changes the game entirely, making eth mining with this thing +13 old school polaris GPUs might as well kill newer nvidia released in hours (mainly because they are spiking low in price and the fact the acorn removes a lot of the power consumption?)

Forgot to add, would it support those 1 to 8 multipliers, cause many have m2 only in their limited gaming desktops.

I think the main restriction will remain to be motherboard support.

Thank you very much. It is helpful and will get back to Squirrel support with this link.

As these are not out yet we can not promise anything with 100% certainty, but at this time we see no reason why the new generation of GPUs wouldn't work with Acorn.

I take that back.  The A3 doesnt work on nicehash either. Let me rephrase that, I've never seen a software miner not work with nicehash   The code to implement (extranonce subscribe) is pretty easy to implement:  https://www.nicehash.com/help/patches-for-extranoncesubscription

So much to test in the next month.

-Bcu1525
-Acorns
-Nvidia 11xx
-Follow 10xx trending price after 11xx series will be available on the market


Rather simple from the outside, but from a ''new'' investor, basically, we have to do the following work ;

1. Test BCU1525 on different algorithms, and calcs our daily income into the differents algo while taking in consideration the 5,000 units that will slowly get online and adjust the profitability to what we will get after they are all online. The hashrate will remain very good, but the profitability will go down a bit.

2. Test acorns on differents Nvidia/Amd hardware and analyse what it's best ; more gpus, or adapt our actual gpus with acorns that yes, we can get insane hashrate for the electricity cost, but we would need minimum 1 acorn per gpu. ( can go to 315$ USD + shipping + customs fee ). It feel kinda obvious that on the long term road it could be the way to go, but it would depend of the profitability differences of getting our whole mine equiped with acorns vs taking that same investment on bcu1525, what is the best hmm.. ?

3. We are not done yet. Nvidia will come to the game soon enough with their new 11xx ( or 20xx, whatever ) series. We have to take into consideration that 10xx series will get down in price, and we will have to test/see research the actual hashrate those cards can handle, and how they can be compatible with the acorns.

------------

Guess we will be busy with some new toys to play around.