You can pool mine without an account, although you are putting trust in whichever exchange you have your rewards going to - I personally don't love that idea. So I'm attempting to solo mine my first accounts (at the moment the reward for a block is 100PASC + 5 accounts linked to your private/public key). Unfortunately I only have a single GTX980 - Judging by the math of the online pascal calculators, I think its going to take me about 3 months of mining at 4-500MH/s before I see my first block. (unless i get real lucky)
I managed to get the CUDA ProxyMiner working - as shown in some of the videos, it comes in two parts, a java proxy applet which talks to the Wallet software and then a miner.exe for each GPU you want to mine on. You can get the CUDA (Nvidia) or OpenCL (AMD) miners here:
https://github.com/Vorksholk/PascalCoin-CUDA/releases
https://github.com/Vorksholk/PascalCoin-OpenCL/releases
1a) Enter the IP of the machine running Pascal Coin Wallet (127.0.0.1 if its on the same computer)
1b) Enter the TCP port setup in Wallet (default 4009)
1c) Enter the number of GPUs to be used on that computer
1d) Enter the miner name (this i'm not 100% sure on - I just matched it to the value I put in the Wallet Options window under "Miner Name" - from my understanding this has to be exactly 8 characters)
PascalCoinCUDA_ProxyMiner_sm52.exe p0 d0
PascalCoinCUDA_ProxyMiner_sm52.exe p0 d1
etc...
With CUDA you'll see several ProxyMiner.exe files with smXX numbers at the end - this relates to the video card support required: I found this handy reference to explain the sm numbers http://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/ - scroll down a little - pick one that looks like it matches the card generation you are targeting.
SM20 – Older cards such as GeForce GT630
SM30 – Kepler architecture (generic – Tesla K40/K80)
Adds support for unified memory programming
SM35 – More specific Tesla K40
Adds support for dynamic parallelism. Shows no real benefit over SM30 in my experience.
SM37 – More specific Tesla K80
Adds a few more registers. Shows no real benefit over SM30 in my experience
SM50 – Tesla/Quadro M series
SM52 – Quadro M6000 , GTX 980/Titan
SM53 – Tegra TX1 / Tegra X1
SM60 – GP100/Pascal P100 – DGX-1 (Generic Pascal)
SM61 – GTX 1080, 1070, 1060, Titan Xp, Tesla P40, Tesla P4
SM62 – Probably Drive-PX2
SM70 – Tesla V100
For CUDA you need an nVidia driver with at minimum CUDA 7 support - so if you've been dicking around with your drivers for Ethereum Mining like I have, it won't work with the recommended 347.52 Driver used for faster Ethereum Mining in Windows 10. FWIW, I'm running the latest driver and when it was working best, was getting around 550MH/s on a non-overclocked GTX 980.
Once I get my first account, I will likely switch back over to pool mining, particularly because I don't really have a proper rig for mining, and I would get ongoing PASC while I mine, rather than play the lottery.
Let me know if you have any further questions - hopefully this is enough to help you get going.
PS: Once you get it going, you'll see the miner say "Found nonce" and the proxy say "GPU X submitted a share", unless I'm mistaken, this is primarily useful in pool mining and is used to help Pool servers assign rewards for progressive work. The nonce found will have a chance that its a block, but is subject to much lower difficulty than the current block difficulty, so mostly, its just a progress ping to the pool server. At the same time this happens, if you happen to be looking at the logs of the Wallet, you'll see the message "Error: No valid job found with these values! (Perhaps prior block job or old job)".
Its a little disconcerting, so I dug through the source code of the miner and the wallet and I'm quite sure its just the wallet telling you the share nonce wasn't a valid block. Here's hoping I'm right, and I haven't been mining for 3 days for nothing ;)