If you have the hardware, you could potentially set up a virtual machine on your Mac. But this is not ideal because you can’t run the hardware you really need to run on a Mac to properly support the gpus.
Just spoke with Wolf0, he says the client could potentially be cross platform Linux/Windows.
I’ll have to check exactly what hardware is running on my mac, but I got the alpha version of zcash running pretty well on a VM copy of linux.
Alternatively, are there any guides to buying/building PCs optimized for mining?
In for 1 btc, will need more concrete details.
I’ll contribute 1 BTC but hope we can get support when it doesn’t work for us, like I have a machine with 6 XFX 380 graphic cards in one mainboard.
1 BTC also here for the crowfunding
1 BTC also here for crowdfunding
I am interested, and can put in 1 BTC.
Interested in GPU miner will put in 1 BTC. Need Linux and CUDA support.
I’ll commit to 1 BTC if it exceeds 3x a CPU’s ability with a $300 GPU. Linux-based for me.
It should be tested against the testnet. The multiple by which it exceeds a CPU with 1600 MHz DDR3 will be equal to the number of blocks it can average per hour times the average time-weighted difficulty divided by 100, preferably after having collected at least 30 blocks. For example, if it collects 30 blocks in 3 hours when difficulty averaged 150, it is 30/3 x 150 / 100 = 15 times better than a decent CPU. A good way to get the difficulty is to take the average of the peaks and valleys, or 1/2 the peaks or 2x the valleys. Currently the network is averaging difficulty 250 (2 day average). On the testnet, when you see a difficulty=1 you don’t use “1” in the average because it is actually being kept a the previous difficulty value.
The included benchmark solver will probably not be good for this. I haven’t been able to correlate blocks gained with the benchmark in a linear fashion for even a single thread.
Which GPU platform are these two developers working on? I mean nVIDIA or AMD.
Ampy said “It is an opencl implementation, it should work on all opencl platforms. That said, I am only testing it for AMD.” And I know wolf0 has been testing with AMD as well.
@zawy You should jump onto the Zcash slack mining channel and ask wolf0 and ampy these concerns. They can answer you directly there. They are currently working independently, and also in competition with others. Until we have produced enough interest, they are very hesitant to give any specs or comparisons at this time. What we do know, however, is that they are very good at what they do. And ampy has hinted that it is safe to assume the speed up will be at least 4x, but he won’t expand on that. If, for example, the speed up was even more noteworthy, than zcash devs may take a particular interest and attempt to thwart gpu development efforts.
I’m in for crowdfund. 1btc from themav
Everyone, as we are discussing the expectations of the client, I need input from interested parties.
Many have specified “open source” as a criteria. Will you please define your expectations? For more information, read the zcash slack channel regarding this discussion.
Quoting wolf0:
“@coin_artist ask them what they mean by open source, and what their goals are. because most people’s goal is trustability and people being able to review the code - which is met by the host code being open. The device code no one needs to verify its safety.”
Input appreciated.
If it works proportionally as well on Nvidia, I’m ready to contribute 1 btc
I want you to be the point of contact.
They may sell it to only 1 party because that will be where the money is. Zcash could make their own GPU miner for the public, or be the highest bidder for wolf0 or ampy. They need to do this because they want it to be widely distributed. If they’ve failed already in being able to have it widely distributed on cell phones and PCs, they need to make the most of it. Keeping ASICs out of it seems unique enough, as I understand it.
wolf0 and ampy could put the copyright up for sale on Ebay and let the highest bidder win, along with an NDA and non-compete agreement. They would need to advertise the results with a few specific GPUs, i.e. how long it takes to get 50 blocks on testnet with each GPU and what the difficulty was during that time period.
Zcash could partner with those posting here, matching the bid, then there would be a commitment to keep the miner between us for say 6 months or 1x return per GPU ($500 BTC + $300 GPU) accepting $1 per coin as the return, whichever comes first, then Zcash would release it. This could soften concentrated mining at startup, let us profiteers be guinea pigs for improvement (wasting money on wrong GPUs and before the code is optimized). I see several here including an “open source” condition which is quite nice of them. This is a compromise between all: the programmers, Zcash, the community, and those of us wanting to make a risky bet.
Or we just agree on when to make it public, if Zcash stays out of it. But I do not think we have enough money. It depends on how much they want to take a loss for charity.
Concerning the benchmark, if they can get it to solve in an average 15 seconds per single core, and can be sure it will scale linearly per core due to high on-board RAM, then > 5x CPU is here. Otherwise, they need to run it on the testnet We need to know if they are using a $300 GPU or a $1000 GPU.
There’s no point in a $1000 GPU doing only 4x better.
Like I said, I’m up for up to 1 BTC… As long as the miner is for both Linux and Windows, if that’s possible. 
I’m in for the crowdfund at 1 BTC.
My interpretation of the 3 who said “open source” is that they prefer us all to finance it, but then give it away to the public. I would prefer it to be just an executable given to those who contribute. If this is not acceptable, then we could say it’s just among the contributors for 6 months, then the coders can release it as an open source project.
I plan on GPU mining zcash, and want to contribute - 1 btc ready. Open-Source software code. Allow the community to freely access it. ROI will be enough just by mining early in the game. I look at it as unhealthy to put up barriers for other people to join in at start.