I can't really make out the configuration, but is this roughly 2/3 of an S7, with the ATX PSU on top and two fans? The 90 chips sounds like two boards. The power requirements and hashrate don't quite fit the "two thirds of an S7" theory, but I'd be stunned if they actually engineered another board for the BM1385 chip at this late date.

As packaging goes, it's pretty obviously a "paste up job" of existing parts (except for the top PSU mounting bracket).

I know I will pass, though I don't think Bitmain will miss my non-purchase.  

@Tupsu: Why do you think this tells us anything about the S9?

Respectfully, I must disagree. There are really only 4 items that matter in terms of purchasing an ASIC based miner, and neither geometry size (aka nm) or AsicBoost matter at all. The items I see are:

1) What does it cost?
2) What is the hash rate?
3) What is the power consumption in Watts?
4) When can it be delivered?

I guess I would also have add are there any special cooling (does it need to be -100C), or electrical supply requirements (i.e. does it need 400Hz 37 VAC as a crazy example).

The internals of how they did it don't matter one whit to me.

Just my opinion, which I expect is fairly common.

If the SP50 existed in any quantity (say 5 or more), and BTCS was running them, why didn't they just show them on one of their videos? At the present time, there are more pictures of "Bigfoot" than there are of an actual hashing SP50. It would only seem the prudent thing to do given their investment in Spondoolies.

I really liked Spondoolies, it's just a shame that things fell apart for them, including the SP50.

I will echo Jonny's sentiments here. Furthermore, my experience is that you need to wait a few hours to see where you actually end up in terms of hashrate and power consumption.  I have tried a setting of 100, and several hours later it was still well over 2.8TH. The first 15 minutes look good in terms of power and fan speed, but later in the day the fan has ramped back up, and the hash rate is way higher than I expected.

This is quite unlike the SP20 in terms of controls and the ability influence efficiency. The efficiency of the Avalon6 can't be changed via cgminer settings (IMHO).

The last 600W miner that Bitmain supplied was the S5. They haven't tried to do some kind of S7-  or "S7 lite". I think they have lost interest in "down scaling" their miners. I'd love to be proven wrong. 

Brace yourself, this one's going to hurt......

Bitcoin Difficulty:    178,659,257,773
Estimated Next Difficulty:    195,059,596,071 (+9.18%)
Adjust time:    After 7 Blocks, About 1.0 hours
Hashrate(?):    1,493,660,864 GH/s
Block Generation Time(?):    
1 block: 8.9 minutes
3 blocks: 26.7 minutes
6 blocks: 53.4 minutes
   
Updated:    23:0 (20.6 minutes ago)

The current number of transactions in a block, times their BTC amount is just minuscule. I think if you were to look at how many blocks that carry transactions (some don't), you'll find that they are maybe 1% of the size of the 25 BTC reward there now. After the "halving", transactions will still be a mere pittance compared to the 12.5 BTC reward.

The total mining infrastructure of today is 99% focused on the block reward in the blocks. Transaction fees are strictly a tiny bonus.

Is there anything more "real" or substantial on the S9? I am skeptical of it's existence at this time.

For what it's worth, nextdifficulty.com has been suggesting an increase of 9.6% for several days. I thought for sure it was way too high.........

I guess they got their signals crossed and turned ON 150PH. Crazy spike on bitcoinwisdom:

Bitcoin Difficulty:    178,659,257,773
Estimated Next Difficulty:    191,583,081,271 (+7.23%)
Adjust time:    After 374 Blocks, About 2.3 days
Hashrate(?):    1,433,575,801 GH/s
Block Generation Time(?):    
1 block: 9.0 minutes
3 blocks: 27.0 minutes
6 blocks: 54.0 minutes
   
Updated:    17:25 (21.5 minutes ago)

Doesn't look to me like a "gentle landing" near 0% like last time.

Whoever is responsible for "Switching off the 150PH" missed the memo before the weekend.

Bitcoin Difficulty:    178,659,257,773
Estimated Next Difficulty:    188,771,332,059 (+5.66%)
Adjust time:    After 536 Blocks, About 3.4 days
Hashrate(?):    1,411,995,799 GH/s
Block Generation Time(?):    
1 block: 9.2 minutes
3 blocks: 27.4 minutes
6 blocks: 54.9 minutes
   
Updated:    17:30 (5.1 minutes ago)

It would seem that you might have to dial down the frequency depending on how good an S3 board you. Wasn't the biggest difference between the S3 and the S3+ the range of usable frequency (i.e. the S3+ being higher)?

You didn't mention the kind of Antminer's, but I would also encourage you to consider one or more of the Gecko Science Compac miners. Very adjustable, way faster than and Antminer U2, and very stable. Antminer U3's have a poor reputation for stability.

Just my two cents.

While it's sad, disappointing, and frustrating, this is as much self-inflicted as something evil on BitFury's part. They never asked for money, and then for whatever reason their plans change. That's life and all the hype that folks laid on top of it along the way, wasn't the fault of BitFury.

In this case, BitFury is NOT Butterfly Labs.......

One minor cautionary note: Since the I/O board only has 3 connectors on it, there is a small chance that there could be a problem with any of the other 6. I would bet that Bitmain doesn't go to the trouble of trying to test the 6 un-populated socket positions. Hence if you solder the connectors down correctly and there is a problem with one, it's unlikely that Bitmain will send you a replacement.

While it should work, and is likely to work, don't expect a guarantee.

So the spoof is the .co -vs- the .com domain?

Very complete replica........

I am not seeing this at all. Is there a chance you missed a zero? Looks to me like 4TH (not 40) produces 0.01126 BTC/day. That suggests that 40TH would be much closer to .1 BTC/day.

I am just using bitcoinwisdom as my simple calculator for this. It also fits with experience with my 3.4TH Avalon 6. That doesn't get anywhere near .1 BTC/day.

I think the more accurate answer is about 400 TH.

While it would be nice to see 150PH "switched off" today, I am thinking that's not going to happen. I think this will turn into a "leg up" for difficulty that won't go away for quite a while (if ever).

Just my kinda pessimistic outlook.

+5.0 = alh   (I think this will fit)

No thank you......