BFL_Josh 2 days ago.
I have no idea where this ludicrous "ground plane issue" rumor got started, but whoever started it is out of touch with any sense of reality. There is no ground plane issue on the PCBs. The hold up right now is the test rig and debugging it... since that is what we are doing at the moment, there's nothing new to update. Once the test rig issue is resolved, we will resume the process going forward. I am loathe to detail the process at this point, it is just fuel for the trolls who like to pick apart every word type and analyze it for conspiracy theories. I will update when I have an update that's worth posting. Smaller updates I will post in the chat, and thanks to the magic of the grrnbrg they will make it to twitter. At this time, I wholeheartedly endorse grrnbrgs unofficial twitter feed.
Just as an unrelated side note, it would take approximately 24 hours to get new PCB's made if we had to redesign them for some reason. Depending on whatever would need to be done to the PCBs, the actual design work would take two or three days typically, unless it was a complete redesign and there would be basically no reason in the world to do that at this point. I don't know why anyone would think 4 - 6 weeks for a PCB redesign, that's just crazy. We have gone through many, many board redesigns since this process started, I've only pictured two or three of the 6 or 7 revisions we've had throughout the process. Basically, we could have a redesigned board ready to go by the time the chips arrived at the facility, that has never been a sticking point.
I haven't been posting much because I am very busy with other issues related to getting these chips out and making sure everything is ready once they arrive here in KC. I will make time to sit down and go through some of the threads here as soon as I can, but right now I am focusing on all the details for shipping. The daily DDoS doesn't help, but I've mostly gotten that automated at this point, so it's no longer that big of a concern or time suck thankfully.
I stand corrected, with the new 10 layer PCBs it takes 48 hours to make a new board from scratch. But I digress... as I said in the shoutbox, we think we've found the problem with the test rig and the ASIC team has managed to get the chips spinning with some hand modifications to the test rig and some of the engines were even running at 500 MHz. They are going to finish diagnosing and then get new hardware for the test rig, so we should have something demonstrable by early to mid next week. Should be pretty smooth sailing after that.