As a newly started company, with such a modular and advanced product, I dont think their target is that broad yet, Its easier for them to sell the first batches to the more hardcore miners, and let them promote and handle most of the q&a for the nicehash guys

Whattomine calls it cryptonight v7 also, but it is monero chain fork 7, that changed from cryptonight (v0) to cryptonight v1 which adds a few operations to the main loop and the setup phase to kick asics off the network.

This is the last time i will say it, what acorn can mine is the current monero fork monero7 using cryptonight v1, the one where you dont fight asics. It does NOT mine the old cryptonight (v0) variant that asics are all over, and it does not mine cryptonight heavy.

I dont see why you need gpuhorder to confirm this, sqrl site clearly says cryptonight v1, and that is whats really driving monero atm, go google it, and you will find you that monero use cryptonight v1, cryptonight v7 does not exist. Ask anyone that knows the technical details of monero they will say its cryptonight v1.

Go read https://coin.fyi/news/monero/cryptonight-v7-vs-v1-a-nitpick-request-to-use-the-correct-algorithm-name-8b8o05

Its a preorder........

Guys please do a little reading before writing shit inhere man, takes 5 minutes to go research the shit yourself, then you would find out that what acorn can mine is monero, cryptonight v1 is the current monero algo, its a branding problem, this fork was the 7th fork on the monero chain, therefor called monero7, but its cryptonight v1, okay ?

Youre right that is an attitude associated with crypto that makes no sense for me also, But for me startup hickups like a webshop that aint working properly under full load is something ive seen manny times before, and mostly outside crypto. So its not like im saying "hey this is crypto what do you expect."

Give them a chance, there is good people behind this project, dont shoot it down before its launched, come august and ppl dont get their stuff shipped then you can start.
Start up hickups is to be expected also outside crypto.
If you dont want shit to end bad, then chill down, and wait.
Its simple if you keep telling good ppl they are bad, they might some day go bad, since "everyone" says they are bad anyway.

And for comparing this project to bitmain, shame on you..

Thats because those are the 2 algo's you get the most bang for your bucks on, so they have prioritized them

In case you want 1:1 you buy a nest x4 2 m.2 to pcie x4 converters, and some pcie x4 risers, and you can have 2 gpu and 2 acorn on the nest x4, and again due to the dma engine all high bw traffic from gpu to acorns and back should he handled on the nest x4, its not that hard to imagine what kind of setups you can do, but i would most likely have bought 10 times more acorns if they could confirm that this is the case.

Im not trying to decide anything just stating facts, if you go dig arround here you will also see it mentioned that standalone mining is not that profitable on them, You can see by looking for example at the tables, that for now primary focus is on lyrarev2 and x16r.

Because thats where you get the most boost and therefor profit with this kind of setup.

I dont get that did 2 orders with no problems.
They are straight forward and clear with the performance numbers, but you might have a hard time wrapping your head arround their numbers.

What i think is really shitty is that they have next to no information on the nest cards..
They have an onboard dma engine, so in theory you should be able to boost 4 gpu's using 2 acorns on a nest x2g card, IN A x1 motherboard slot. But noone want to confirm or deny.

For solo fpga mining youre the wrong place go buy a bcu1525, acorns are generaly to small to fit solo mining without to manny dirty tricks.

Youre verry limited when it comes to pcie speed on h110, but a nest x2g solution would work.
Ie put 2 acorns and 4 vega's on a nest x2g.
If you want full boost of vegas, then something like nest x4 should work, 2 acorns and 2x m.2 to pcie x4 adapters and then have the vegas there running x4.

But with such a setup you need to trust that the nest boards dma engine works as they say, and eliminate the need for high throughput data comming to the motherboard.
It might be better to wait and see when the first batch is shipped and the reports gets in.

Sure they can pay the import tax, I dont know how it works for your area, but here they could just send all the shit thats going to europe to some warehouse in europe, pay the import tax themself for it all, and then ship it to customers from there.

B250 Boards come with x16 slot that is on the cpu, thats your only hope for acorn, you can fit 4 acorn there, but your gpu's is gona be on the chipsets pcie lanes, so it have to go over dmi, dmi like a 4x pcie 3.0, so 2 acorn will be the limit, so you can run a 2:8 setup. Given your gpu's can run pcie 3.0 you can run a 2:4.

The other solution is Nest X2G, if set up right, the onboard pcie switch and dma controller will handle the high bandwidth data, and it can be run of a 1x slot. But youre still limited to whatever max gpu your board have.

So for example with a Biostar TB250 12gpu board you can have 3x Nest X2G with 4 gpu and 2 acorns on each.

Due to the pcie x1 3.0 ports on the Nest X2G for gpu's you can not run a 1:1 setup like this. 1:1 on B250 is out of the question

( Got some crazy idea while writing this, to use m.2 slot to pcie x4 converter, then you could use 2:2 on a Nest X4, 2 gpu and 2 acorn all on pcie x4, and have the high data contained there so you could stick it into a x1 slot on the motherboard, but better motherboards is cheaper then. )

In some situations you get a 3x boost with the acorn another gpu would give you 1x boost, and at a massive power use, the acorn use 15W, im sorry to say it but I thnk youre to dumb to see the advantages here.
This definately aint for your standard newbie miner that dont know shit about the inner workings.

I hear the resale agurement a lot, well guess what noone wana buy your shitty gpu's with fans sounding like they are about to die.
You might offload some of them to some idiots that dont know shit, but then youre a shady person.
I see more resale/reuse potential in the acorn, its a xilinx fpga with quite fast a pcie bus, you can reprogram it to do whatever you want, accelerating all kinds of shit. Be that video encoder, data processing, bruteforcing, you name it.

It was refering to the whole NestX2G thing, if youre fine with a 1:2 ratio its the shit man.

Like ive said before, and the talk about this dma bus just confirms it, the nestx2g most likely only need x1, if you just use it to boost its onboard gpu's. Since the data comming in and out to the nestx2g to/from the motherboard will just be work and shares.

The data is most likely so small going from the nestx2g in such setup that you could put 3-4 on a switch, add 12-16 gpu's on them, and then run the whole shit with some shitty soc that has a pcie lane. Depending it can run that manny gpu.

The ideas is endless man, what if you skipped the driver and talked directly with the gpu's, so the host computer didnt even need to know the gpu's were on the nest x2g, that would be cool, hint no limits on gpu's

43:30 This is what im exited about too man

"Without having to go back to the system cpu AT ALL"

Remember its an xilinx fpga, so its not some cheap chinese asic core, you still see manny of their 10 year old fpga's arround working.

Did some testing on TB250-BTC PRO the motherboard is not as starved for lanes as you would think.

CPU PCH ( 1 gpu here running full x16 8 GT/s per lane. )

01:00.0 VGA compatible controller: NVIDIA Corporation Device 1c03 (rev a1)
                LnkSta: Speed 8GT/s, Width x16, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-

CHIPSET PCH ( 3 gpu here running x1 8 GT/s through risers. And then some extra cards just to show that running more ports is posible while doing 3 ports gpu at 3.0, but only had extra 2.0 cards :/ )

0a:00.0 VGA compatible controller: NVIDIA Corporation Device 1b81 (rev a1)
                LnkSta: Speed 8GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
12:00.0 VGA compatible controller: NVIDIA Corporation Device 1c03 (rev a1)
                LnkSta: Speed 8GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
13:00.0 VGA compatible controller: NVIDIA Corporation Device 1b81 (rev a1)
                LnkSta: Speed 8GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-

02:00.0 PCI bridge: ASMedia Technology Inc. Device 1184
                LnkSta: Speed 5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
08:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03)
                LnkSta: Speed 5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
0b:00.0 PCI bridge: ASMedia Technology Inc. Device 1184
                LnkSta: Speed 5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
11:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2116 PCI-Express Fusion-MPT SAS-2 [Meteor] (rev 02)
                LnkSta: Speed 5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-

Now I only had those 2 pcie bridges and 2 sas controllers for pcie cards that are not gpu, but i guess some more pcie 3.0 devices works just as fine.
Downside is that with mining mode (in bios) enabled pcie is limited to 2.0 for some stupid reason most likely to not cause problems with shitty risers and otherwise default bios settings, and with mining mode disabled its 3.0 but it can only post with 4 gpu's.
If Biostar would provide a bios that allowed pcie 3.0 while in "mining mode" you could use this board for a 2 gpu : 1 acorn setup.
4x acorn in CPU PCH x16 port, 8x gpu in the chipsets x1 pcie 3.0 ports.
I have already written to biostar asking if this is something they are willing to do.

With the roi time atm I would rather use what hardware I have for this.

If your goal is just to make a setup with as manny acorns as posible running solo mining then you most likely wont need 4x for them each since they should not push any big data to the motherboard.
Futhermore even with gpu's in the mix using the nest2g cards the high throughput should stay on the pcie switch the nest2g card have and not reach the motherboard, but i guess noone know since i stated this multiple times but getting ignored

Well incase the onboard pcie switch on the nest2g card makes the acorn talk direct with the gpu, so the high throughput data never reach the pcie bus of the motherboard only work and result, and the nest2g then can run of a 1x port, then I can come up with some pretty faar out ideas

Site says nothing about bifurcation, but it imply that the board does it since it says any pcie 2.0 x8 or higher port will work.

About the gpu's, most likely their comunication with the fpga's goes through the onboard switch nest2g have only, so the motherboard does not matter in this case, I dont understand why Nest2G would even require the x8 2.0 motherboard port, cant see what the high bandwidth there will do, other than for cross communication if you have multiple nest2g. But that makes no sense.

But thats how it is with a product like this, there is so manny options that they will need to hit the market before all use cases will be tested.

EDIT: Hmm thought about it and yeah cant see at all what high data throughput there would be on the motherboard from having multiple nest2g with 2 or 4 gpu's connected to the onboard ports, they should each talk with their gpu's over their onboard switch and there should be no cross communication over the motherboard since each nest2g will have its own acorn and gpu's, the data comming in and out of that nest2g pcie interface to motherboard should be just work and results. And therefor you should be able to just stick them in a 1x port.

This is all theory!!!!!!!!!!!!
But what do you think ?

Inetresting, does the Nest X2G also do that ?