Acorn M.2 FPGA based GPU Accelerator

[grendel25]

guys please read the drawfminer thread


https://bitcointalk.org/index.php?topic=3628532.0


it ended in a disaster.

be careful is all I have to say.

It looks familiar to dwarf also:

Name:    GPUHoarder
Posts:    206
Activity:    84
Merit:    23
Position:    Member
Date Registered:    March 01, 2018, 05:55:00 PM
Last Active:    June 30, 2018, 05:16:23 PM

No it doesn't.  As someone else said: real face, real YouTube interview with his face, real office address, real project.  Dwarfminer sucked but there were warning signs.

Bitconnect had real faces, realy YouTube interviews with "faces"... but it's hard to make comparisons on office addresses and real projects.  Look, at this point in crypto everything has to be treated like a scam until it 100% is not.  Even Bitmain is widely criticized for scam behavior for the seemingly obvious "secret mining" or "testing"  or whatever it is that they deny "they" have done... sure, maybe "they" didn't... "maybe" it was their subsidiary or some other way of masking whatever the pre-mining before release of the product. 

I have watched the videos.  I agree with most that they all 'seem' very genuine.  They 'seem' like the most awesome people in the world that I would love to hang out and chat with about the awesome things they are doing and even be part of it by purchasing some M.2 boards... maybe even an FPGA once this market turns around and some love starts flowing again to the alts.

BUT... anyone who has been in this sceen since 2013 and even longer has just seen WAY too many scams to trust even the most seemingly benign of projects.  And so, we just have to wait and see.  I think it's more than likely that with this Acorn and Mineority venture that early adopters will be rewarded.  BUT... I could also be very wrong and there could be delays and disgruntled early adopters.  heck.. I might even turn into one... lol

And consider this... Consider the business value aspects for Acorn, Minority, Allmine (scary name by the way... did anyone reconsider the name "allmine" might be intimidating to early adopters?  "Allmine" as in "allyours"? ehh..)  so yeah... business value is huge.  There is an opportunity to make mining more efficient and profitable so of course Acorn should do well here. 

Business value for more profitable mining... Check, got it.  Easy.  Know what else has business value?  Security.

Businesses that assure consumer security also stand to make A LOT OF MONEY.  Of course, Acorn, Mineority, etc could include aspects of secure payment options such as PayPal to capitalize as Dwarf Miner did and then did not as an obvious bait/switch.  But I'm not just talking about secure payment options.

Reden has an anti-scam mindset that could be interesting.  I'm sure there could be some other coin projects on a similar plan.  But there is a lot of business value here.  There are all sorts of ICOs indicating business values that I've seen: Entertainment, Travel, "uberization" of various things... But these are the things that have my attention lately:

1) Mining efficiency improvements  (basically people selling the pick axes as the old adage goes: "it was mostly the pick axe sellers that made money during a gold rush")

2) Automated and Secure shared masternode services where people can quickly begin generating passive income despite only acquiring some percentage of a required collateral... A LOT of people are interested in this

3) Security, anti-scam, escrow-services, improved payment options... Reden coin first comes to mind... I'm sure there are others.

It has been a few days since I poked around for Acorn news... are we getting close to seeing performance results?  I'm about 30 minutes into the GPU hoarder interview with Seth Estrada and I really love it.  Looking forward to the coming days/weeks/months/years.  Wish I could quit my day job and just do crypto stuff... maybe someday hehe

[1714795235]

1714795235

[1714795235]

1714795235

[mfurman]

Wow lots of activity over here - sorry for not being more attentive to the forum post.

As some have mentioned, there is an extremely active (> 2500 members?) discord focused on FPGAs and a ton of information and updates have been flowing there. Let me try to address the bullet points.

I have a simple request:  Can you and Senseless record a Skype or other video conversation where you talk about your project, timetables, your backgrounds, and how you came to team up on this?  I just watched your Seth Estrada interview and his ADD drives me crazy.  I am on Discord but the signal to noise ratio is insanely high and I value my time.

If the two of you can just record an interview and then post in on YouTube, you can point people to that instead of the answering the same questions over and over and over on Discord.

Just script out some broad topics ahead of time and take it from there.  This is a really, really interesting development and I'd rather hear the straight dope from the two of you without various YouTube personalities in the mix.

Thank you.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

[i_like_bits]

Businesses that assure consumer security also stand to make A LOT OF MONEY.  Of course, Acorn, Mineority, etc could include aspects of secure payment options such as PayPal to capitalize as Dwarf Miner did and then did not as an obvious bait/switch.  But I'm not just talking about secure payment options.

They already tried.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

I don't understand why you people can't wait. You'd rather that they'd just make up numbers and tell you what you want to hear, instead of getting everything squared away, and presenting you with accurate numbers/benchmarks. I actually appreciate a company that doesn't rush out data/info without verifying the accuracy. Nobody has to accommodate your schedule. Nobody has paid any money for these accelerator's. Having no patience is the reason there is so many scams on this forum.

[bittawm]

i bet this is legit,

Dwarf miner was a con from the word go

[philipma1957]

Businesses that assure consumer security also stand to make A LOT OF MONEY.  Of course, Acorn, Mineority, etc could include aspects of secure payment options such as PayPal to capitalize as Dwarf Miner did and then did not as an obvious bait/switch.  But I'm not just talking about secure payment options.

They already tried.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

I don't understand why you people can't wait. You'd rather that they'd just make up numbers and tell you what you want to hear, instead of getting everything squared away, and presenting you with accurate numbers/benchmarks. I actually appreciate a company that doesn't rush out data/info without verifying the accuracy. Nobody has to accommodate your schedule. Nobody has paid any money for these accelerator's. Having no patience is the reason there is so many scams on this forum.

but he did rush to start the thread and has missed his dates.

I never said anything negative about him everything was true.

he said  mid june for preorder >>>>>>>>>>> missed this date
he said late june for shipping >>>>>>>>>>>> missed this date
he mentioned showing proof with a video >>>> backed off this statement

he has done none of the above.

let us pretend he is fully 100% honest and will deliver every day of delay HERE  means we get closer to gear like this


https://bitcointalk.org/index.php?topic=4553947.msg41248146#msg41248146

So I was looking forward to this working right now as I type instead it has been hype.

[GPUHoarder]

Businesses that assure consumer security also stand to make A LOT OF MONEY.  Of course, Acorn, Mineority, etc could include aspects of secure payment options such as PayPal to capitalize as Dwarf Miner did and then did not as an obvious bait/switch.  But I'm not just talking about secure payment options.

They already tried.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

I don't understand why you people can't wait. You'd rather that they'd just make up numbers and tell you what you want to hear, instead of getting everything squared away, and presenting you with accurate numbers/benchmarks. I actually appreciate a company that doesn't rush out data/info without verifying the accuracy. Nobody has to accommodate your schedule. Nobody has paid any money for these accelerator's. Having no patience is the reason there is so many scams on this forum.

but he did rush to start the thread and has missed his dates.

I never said anything negative about him everything was true.

he said  mid june for preorder >>>>>>>>>>> missed this date
he said late june for shipping >>>>>>>>>>>> missed this date
he mentioned showing proof with a video >>>> backed off this statement

he has done none of the above.

let us pretend he is fully 100% honest and will deliver every day of delay HERE  means we get closer to gear like this


https://bitcointalk.org/index.php?topic=4553947.msg41248146#msg41248146

So I was looking forward to this working right now as I type instead it has been hype.

I have to admit jumping back to this thread after spending so much time in the discord is very interesting, since most of these topics have been covered many times there as well as more granular insight into all the projects going on.

As to legitimacy of my self/company, I’ve done about all I think I can do there at this point. Regarding the Acorn, the legitimacy of that specific project  is set to come soon, but I will remind you we aren’t taking anyone’s money right now either.

If I wanted to scam people I probably would be more anxious to start taking money instead of missing a few of my ambitious deadlines while striving to make sure everything was done the right way.

Production planning has been a major endevour. Interest is higher than we initially thought, and securing all those individual components up front so we can be confident on timelines  is a tremendous effort. Little things happen -  a oscillator chip you were promised in volume for July delivery gets updated as backordered till October and you have to find an Alt, etc. We have solved most of those problems now.

Website going up next week, pre-orders a few days after to avoid the “no time to read” problem. Our goal is to meet demand within Q3.

I don’t think I’ve dodged any questions, but people sure do like to make up their own answers. That’s why I’ve been careful to not release a bunch of partial information that gets misconstrued. Let me hit the big ones though.

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).


Edit: I also want to clarify that while Tom (@senseless) and I have collaborated on the 1525 and are generally doing some work together regarding FPGAs, he and AllMine have no involvement in the Acorn project, which is entirely a SQRL (Squirrels Research Labs) project. I’ve noticed some confusion around that.

[R0mi]

I have to admit jumping back to this thread after spending so much time in the discord is very interesting, since most of these topics have been covered many times there as well as more granular insight into all the projects going on.

As to legitimacy of my self/company, I’ve done about all I think I can do there at this point. Regarding the Acorn, the legitimacy of that specific project  is set to come soon, but I will remind you we aren’t taking anyone’s money right now either.

If I wanted to scam people I probably would be more anxious to start taking money instead of missing a few of my ambitious deadlines while striving to make sure everything was done the right way.

Production planning has been a major endevour. Interest is higher than we initially thought, and securing all those individual components up front so we can be confident on timelines  is a tremendous effort. Little things happen -  a oscillator chip you were promised in volume for July delivery gets updated as backordered till October and you have to find an Alt, etc. We have solved most of those problems now.

Website going up next week, pre-orders a few days after to avoid the “no time to read” problem. Our goal is to meet demand within Q3.

I don’t think I’ve dodged any questions, but people sure do like to make up their own answers. That’s why I’ve been careful to not release a bunch of partial information that gets misconstrued. Let me hit the big ones though.

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).

Thanks for the clear answers.

I, for one, have no doubt about your project or you or Tom.

That said, I think a short video on YouTube would really be welcomed and save both you and Tom a lot of time in the long run.

Discord is a massive time suck.  Yes, I have hung out there for some weeks and have seen your answers repeated over and over.  But not everyone can access Discord at work and it is a gigantic time suck.

You guys have pulled this together out of nowhere.  You are the two most interesting people pushing this project and I would like to see you and Tom talk about it on YouTube without the distractions of "live streaming" or what have you.

Thank you.

[mfurman]

I have to admit jumping back to this thread after spending so much time in the discord is very interesting, since most of these topics have been covered many times there as well as more granular insight into all the projects going on.

As to legitimacy of my self/company, I’ve done about all I think I can do there at this point. Regarding the Acorn, the legitimacy of that specific project  is set to come soon, but I will remind you we aren’t taking anyone’s money right now either.

If I wanted to scam people I probably would be more anxious to start taking money instead of missing a few of my ambitious deadlines while striving to make sure everything was done the right way.

Production planning has been a major endevour. Interest is higher than we initially thought, and securing all those individual components up front so we can be confident on timelines  is a tremendous effort. Little things happen -  a oscillator chip you were promised in volume for July delivery gets updated as backordered till October and you have to find an Alt, etc. We have solved most of those problems now.

Website going up next week, pre-orders a few days after to avoid the “no time to read” problem. Our goal is to meet demand within Q3.

I don’t think I’ve dodged any questions, but people sure do like to make up their own answers. That’s why I’ve been careful to not release a bunch of partial information that gets misconstrued. Let me hit the big ones though.

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).

Thanks for the clear answers.

I, for one, have no doubt about your project or you or Tom.

That said, I think a short video on YouTube would really be welcomed and save both you and Tom a lot of time in the long run.

Discord is a massive time suck.  Yes, I have hung out there for some weeks and have seen your answers repeated over and over.  But not everyone can access Discord at work and it is a gigantic time suck.

You guys have pulled this together out of nowhere.  You are the two most interesting people pushing this project and I would like to see you and Tom talk about it on YouTube without the distractions of "live streaming" or what have you.

Thank you.

I am echoing your comments - word for word! Some info is needed and Discord is not accessible easily. We do not want to read (or listen) to lots of noise to to get a glimpse of what is going on!

[gambera]

Businesses that assure consumer security also stand to make A LOT OF MONEY.  Of course, Acorn, Mineority, etc could include aspects of secure payment options such as PayPal to capitalize as Dwarf Miner did and then did not as an obvious bait/switch.  But I'm not just talking about secure payment options.

They already tried.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

I don't understand why you people can't wait. You'd rather that they'd just make up numbers and tell you what you want to hear, instead of getting everything squared away, and presenting you with accurate numbers/benchmarks. I actually appreciate a company that doesn't rush out data/info without verifying the accuracy. Nobody has to accommodate your schedule. Nobody has paid any money for these accelerator's. Having no patience is the reason there is so many scams on this forum.

but he did rush to start the thread and has missed his dates.

I never said anything negative about him everything was true.

he said  mid june for preorder >>>>>>>>>>> missed this date
he said late june for shipping >>>>>>>>>>>> missed this date
he mentioned showing proof with a video >>>> backed off this statement

he has done none of the above.

let us pretend he is fully 100% honest and will deliver every day of delay HERE  means we get closer to gear like this


https://bitcointalk.org/index.php?topic=4553947.msg41248146#msg41248146

So I was looking forward to this working right now as I type instead it has been hype.

Phil, nice videos on the Ethereal capital Blackbox early model. Since they supplied you with a testing ring you seem to trust them.

Now, I just checked their latest 16x FPGA unit and they will include Acorn 215+ in their August batch.

https://blackbox.cat/collections/frontpage/products/blackbox-x16-bcu1525-fpga-system-with-16x-xilinx-bcu1525-fgpas-august-batch

Is Acorn still hype or Ethereal capital is riding the hype? Or they know more than you do?

I'm keeping an eye on this project, it seems interesting.

[philipma1957]

Wow lots of activity over here - sorry for not being more attentive to the forum post.

As some have mentioned, there is an extremely active (> 2500 members?) discord focused on FPGAs and a ton of information and updates have been flowing there. Let me try to address the bullet points.

I have a simple request:  Can you and Senseless record a Skype or other video conversation where you talk about your project, timetables, your backgrounds, and how you came to team up on this?  I just watched your Seth Estrada interview and his ADD drives me crazy.  I am on Discord but the signal to noise ratio is insanely high and I value my time.

If the two of you can just record an interview and then post in on YouTube, you can point people to that instead of the answering the same questions over and over and over on Discord.

Just script out some broad topics ahead of time and take it from there.  This is a really, really interesting development and I'd rather hear the straight dope from the two of you without various YouTube personalities in the mix.

Thank you.

It was actually GPUHoarder in the interview but otherwise I completely agree with you. Seth Estrada's interview style and 85% waste of time on nothing drives me insane. It was such a waste of GPUHoarder's time!
Also, it is time to show some results. We have to plan what to do! I am going back and force between FPGA, Acorns and just getting a couple of more ASIC miners and forgetting GPU mining for good. Decision time for me is the next couple of days. Instead of all those discussions and interviews, we should get some real information!

I spent forty minutes watching it and could not stand the interviewer so I gave up.

It would be a simple thing to film twenty minutes of the gear running with the acorn2 and then without the acorn2.

I spent at least three hours reading discord and not much info.

At this point I know intel optane uses a m2 ssd style device in a m2 slot and it works faster.
I also know ohgoditisagirl ETH enlargement pill does make my 1080ti go from 33 to 50 Mining ETH.

So the idea of an acorn2 speeding up my cards seems plausible.

I will post here on and off.  Plus read here each day.

But a demo video would help you out as a picture helps.

When I do videos on YouTube I do them raw no edits no bs and I do them alone with the gear.

Most all of us are gear heads.
And dont care much about slick production values. Just shit that shows the gear works.

By the way ohgoditisagirl and eth enlargement pill is an exception to slick production not working as it does work as a good sell and as a miner booster.

They are why I think your gear can work.

[KaydenC]

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

Why is Acorn on PCIe 2.0 and not 3.0. It seems bandwidth is a limiting factor, and doubling bandwidth can bring significant improvements.

[GPUHoarder]

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

Why is Acorn on PCIe 2.0 and not 3.0. It seems bandwidth is a limiting factor, and doubling bandwidth can bring significant improvements.

Why aren’t all cars on the road lambos?

PCIe 3.0 Chips are significantly more expensive, this is the right place on the cost/value curve today. But it is being worked on.

[grendel25]

I want to believe.  I mean, I want to put money (at a reasonable price of course) on this as soon as it is available.  But if it were available tomorrow I would not have a clear outline of what is an optimal configuration.  I don't know the best motherboard to buy.  I don't know the best video cards to work in tandem.  idk idk idk...

My least favorite word when it comes to speculative ventures is the word, "soon."  I am sure GPUhoarder would not want to venture any further guesses at timelines at this point.

So... I'll just hurry up and wait.  Good luck  

edit: good to see GPUhoarder answering questions here as well as in Discord.  I lurk in discord too but... idk, I prefer to know a good place where I can easily find the latest/current status updates.

/lurk mode

[CreamyG31337]

You guys shouldn't be so harsh about Seth's interviews -- there's nothing wrong with his style. He's done lots of helpful interviews lately shedding light on the FPGA projects and mineority, plus got a few coin teams to talk about their projects goals, use cases, how they feel about decentralization, proof of useful work, and other interesting stuff. As for the FPGA interviews being light on technical details, that's to be expected at this early stage when bitstreams are being developed and hardware is still being finalized. I'm not sure why you are even asking for a video showing the hardware running when it's been stated so many times that performance metrics are not being made available yet, and the reasons have been given why.

[mfurman]

You guys shouldn't be so harsh about Seth's interviews -- there's nothing wrong with his style. He's done lots of helpful interviews lately shedding light on the FPGA projects and mineority, plus got a few coin teams to talk about their projects goals, use cases, how they feel about decentralization, proof of useful work, and other interesting stuff. As for the FPGA interviews being light on technical details, that's to be expected at this early stage when bitstreams are being developed and hardware is still being finalized. I'm not sure why you are even asking for a video showing the hardware running when it's been stated so many times that performance metrics are not being made available yet, and the reasons have been given why.

I still believe that at least 70% of the time of these interviews are his interruptions, attempts on jokes and other fluff. Such a waste of, quite often, more than two hours of the time. He has good interviewees but wastes their time on pointless chat and his interruptions

[dragonmike]

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).


Edit: I also want to clarify that while Tom (@senseless) and I have collaborated on the 1525 and are generally doing some work together regarding FPGAs, he and AllMine have no involvement in the Acorn project, which is entirely a SQRL (Squirrels Research Labs) project. I’ve noticed some confusion around that.

Can you clarify the algo(s) you are talking about when you mention "MH" in general? I know you specify bit state but personally I have no clue which algo uses which bit state...

Also, some real-life practical examples for people with existing rigs would be very useful - most of us got multi-GPU rigs chugging along - typically with x1 risers - and it'd be good to know what we could expect if we plugged an Acorn into the M.2 slot.

Finally, a maybe a "how-to" guide to extract the max out of an Acorn for an existing rig, considering all the parameters you mentioned, eg. "should we change PCI gen settings in BIOS", or "should we replace one of the 1x-16x riser with a 16x-16x riser on one of the GPUs to get the lift" etc.

[mfurman]

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).


Edit: I also want to clarify that while Tom (@senseless) and I have collaborated on the 1525 and are generally doing some work together regarding FPGAs, he and AllMine have no involvement in the Acorn project, which is entirely a SQRL (Squirrels Research Labs) project. I’ve noticed some confusion around that.

Can you clarify the algo(s) you are talking about when you mention "MH" in general? I know you specify bit state but personally I have no clue which algo uses which bit state...

Also, some real-life practical examples for people with existing rigs would be very useful - most of us got multi-GPU rigs chugging along - typically with x1 risers - and it'd be good to know what we could expect if we plugged an Acorn into the M.2 slot.

Finally, a maybe a "how-to" guide to extract the max out of an Acorn for an existing rig, considering all the parameters you mentioned, eg. "should we change PCI gen settings in BIOS", or "should we replace one of the 1x-16x riser with a 16x-16x riser on one of the GPUs to get the lift" etc.

Very good points. I would also like to add that there seems to have been more emphasis on, shall I say, smaller coins (and newer algos) but I would welcome some clear numbers for, lets say, ethash and other algorithms like it. This is an issue for those who are deciding if GPU mining is still viable or we should all just give up and buy ASIC miners. For the record - I am very happy with Z9 mini so far.

[CreamyG31337]

You should assume that 1x risers are not supported. It's already been made quite clear the acorn is not going to be useful in that situation unless you run it by itself, not talking to the GPUs.
PCI-e v1 should work fine if your board provides 8 lanes to the integrated m.2, or you can use an 8x to m.2 or 16x to a dual m.2 adapter, plus your slave card(s) are in 8x slots. The math is really obvious and simple -- if you want to use half-speed pci-e from what is suggested, then double the lanes. If you use pci-e v3 it doesn't speed up the acorn so that will still need x4, but the video cards could get away with 2 lanes. There's no reason you should ever need a 16x riser to the video card!
I'll try to figure out which algos are which state 256/512 but it's more complicated than that - you should expect a 1060 to get a bigger boost than a 1080 Ti, for example.

lyra2rev2 - 256 bit
lyra2z - 256 bit
neoscrypt - 256 bit
equihash based (zhash etc) - 256 bit
cryptonight based (v7, heavy, light) - 256 bit
phi1612 - 512 bit
x16r/s - 512 bit
phi2 - 512 bit
x17 - 512 bit
tribus - 512 bit
skunk - 512 bit

[ciprianpt]

1. Lift is algorithm specific (some coins have a 512bit state and some have a 256bit state). Each Acorn can provide a MAX of around 30MH of lift to algorithms with a 512bit state, and 60MH for algorithms with a 256 bit state. That is irrespective of the number of GPUs. Power consumption of Acorn at peak is 15W. The limit is PCIe bandwidth, which for this generation of Acorn is PCIe 2.0 x4 = 20Gbps nominal, which yields an effective rate of 16Gbps after accounting for overhead. That math will then get you to the 33MH @ 512bit state, less 10% for safety and underpromising.

2. The number of GPUs to Acorn is extremely GPU and algorithm dependent. X16R  is good with 1:1, or even 1:2 on high end cards if your GPU is in x8. ETH is typically more GPUs per Acorn.

3. The important thing is your PCIe bandwidth has to add up. You can’t use an Acorn in an M.2 slot @ 20Gbps to effectively accelerate one GPU in a PCIe 1.0 riser at x1 (2.5Gbps).


Edit: I also want to clarify that while Tom (@senseless) and I have collaborated on the 1525 and are generally doing some work together regarding FPGAs, he and AllMine have no involvement in the Acorn project, which is entirely a SQRL (Squirrels Research Labs) project. I’ve noticed some confusion around that.

Can you clarify the algo(s) you are talking about when you mention "MH" in general? I know you specify bit state but personally I have no clue which algo uses which bit state...

Also, some real-life practical examples for people with existing rigs would be very useful - most of us got multi-GPU rigs chugging along - typically with x1 risers - and it'd be good to know what we could expect if we plugged an Acorn into the M.2 slot.

Finally, a maybe a "how-to" guide to extract the max out of an Acorn for an existing rig, considering all the parameters you mentioned, eg. "should we change PCI gen settings in BIOS", or "should we replace one of the 1x-16x riser with a 16x-16x riser on one of the GPUs to get the lift" etc.

my thoughts on this https://1stminingrig.com/is-gpu-mining-stronger-than-ever-acorn-m-2-fpga-based-gpu-accelerator/

[CreamyG31337]

You guys can save some money and make a free wire transfer to Squirrels at transfer.xe.com
It saved me about $100 on one BCU, for example. It will be nice when the acorns are available as bank wires are quite annoying for smaller transactions. If you decide to add a riser card or your friend asks you for an acorn a day late it's really not going to be nice paying those fees again. Plus you don't even have to drive to the bank if you do it this way.

Simply sign up, and then create a new transaction, filling in your grand total from the invoice where it says "you want:", after making sure USD is selected. It will calculate the amount in your local currency so you can compare to your bank at this stage.
Continue to the next screen, and select destination country "United States". Take the information from your invoice and fill it in here. For example, but please refer to your own invoice:
Recipient's name: Squirrels, LLC
Recipient's address: 121 Wilbur Dr NE North Canton, OH 44720 USA
Bank account number: 233220xxx (please copy from the invoice!)
ABA Routing Number: 044000037
(other bank details should fill in and match the invoice)
Reference: (Your invoice #)
Reason for payment: "Goods/Services"
Continue through the process to commit to this transaction at your quoted rate.

Then, at your bank website:
Create a bill payment to xe's partner company for your country using the 'pay to' as the name of the bill or payee, and your unique 'client number' for the account number. You don't need to put the reference to quote anywhere. This information will be in a pdf they have on the site and also e-mail to you.

Optional
Call xe and let them know the purpose of the transfer to ensure no hold is placed on the xfer due to any suspicions of money laundering

If your bank doesn't have the required payee to send the money as a 'bill payment', you can upload a voided cheque to xe and they will link directly to your account to debit it, but this will take about a day for them to set up.

My transfer went from my bank to xe (HIFX Canada) in 14 hours. The wire was sent 6 hours later, and payment was confirmed by Squirrels about 40 hours later, which is pretty good for having bank holidays in both countries this week.