Klondike mainboard choice:
Status of the Klondike 16: Rev1 board design review, prototyping, and initial testing is now complete. The Rev1 is successfully submitting shares with one chip using cgminer, though there are some small issues with results. Please follow the thread
https://bitcointalksearch.org/topic/klondike-16-chip-asic-open-source-board-preliminary-190731 for up to date information regarding the Klondike mainboard. Testing of the fully populated board is scheduled to begin shortly. Upon final testing of the fully populated board, final purchasing of components and PCB will begin.
Initial production and assembly will be based on the Rev1 Klondike, which will be used until the Rev2 design has been completed. This decision is based on several factors.
Time to market: The Rev1 is scheduled to be fully tested and ready for production the first week of July. This is a very tight timeframe, and production costs for Batch 1 will increase as a result. This increased cost will not be passed on to the end user. Once Rev1 is complete, Rev2 will likely take at minimum one month to design, order, test, and finalize. This timeline does not take into consideration any complications or design reworks.
Marginal overclocking gain: The overclocking headroom gained from Rev2 does not justify the delay in production caused by increased development times. At the stock value of 282mh/s, the Rev1 design uses 12.8A of the 16A rated capacity of the regs. With proper cooling and airflow the design should allow for overclocking up to 350mh/s per chip using roughly 15A.
Increased cost versus decreased benefit: The Rev2 design should deliver enough power to allow overclocking to the theoretical limit of the Avalon ASIC chip of 450mh/s, but comes with a price. Standard air cooling will not be sufficient to cool the chips at this level. Liquid or oil based cooling solutions must be used, which increases the per-unit cost. Additionally, the board itself becomes more expensive as board component costs increase.
Efficiency and reliability have been the focus of this project from the very beginning. We understand the desire to overclock the Avalon chip to the absolute maximum, but this simply is not an efficient allocation of resources. That being said, we will continue to support BkkCoins and his development efforts just as we have with the Rev1 design, and will likely offer the Rev2 design once finalized to those that desire it. We cannot however, justify delaying entire batches of orders while we wait for Rev2 to be completed.