For the past 12 months I have been getting to grips with PixInsight. As with most astrophotography software, each product you use is different and it takes quite a while to build up your knowledge and get decent results. Luckily, there are plenty of videos and tutorials to help and eventually you get a workflow that works for most, but not all, situations.
As you progress you find that, whilst your basic workflow is satisfactory, it is only a starting point and there are lots of different ways where you can do something different and gain improvements.
And this is where I ran into problems. I started of with a PC that was no more than a basic home use workhorse. This was fine for the basic workflows, other than it just took a long time.
Once you start going beyond the basic workflow the need to experiment and try different things means that you need more immediate results so that you can try alternatives and not just follow a single routine. Unless you are prepared to wait forever, and then waste that time as it didn’t work as expected and then waste more time trying something else, the basic home PC just doesn’t deliver.
PixInsight do provide some help here. You are able to benchmark your PC set up and compare it against other setups. At the time I was using PixInsight Core 1.8.9-2(x64). There were 147 different processor models with up to 50 different setup benchmark results for each processor model.
Out of the 147 processors in this group, my set up was listed as number 146 out of 147. The benchmark score was 2433 against the best of 53237. The time to do the benchmark test 193 seconds against the best of 8 seconds or so.
My solution was clear, get a faster and more capable system.
The next question was how much I was prepared to pay. Whilst time waiting during processing was important I felt I only needed to achieve a factor of x 3 or x 4 in terms of speed to allow me to get the improvements in final quality I needed so that I could try different ways of doing things. A better camera or better filters would probably give me a better result in terms of final quality, but just improving the data quality without a means of getting the best out of this data was not a sensible way forward.
In the end I decided that I would seek to achieve the x3 or x4 improvement at the lowest cost possible.
I spent £350 and this moved me up the benchmark table to position 50 out of 147 with a score of 17202 and a benchmark time of 27 seconds. This was better than I expected.
Was it worth it? The speed in using PixInsight is a revelation. Just opening and closing files and processes etc has gone from a clunky operation to a smooth and reactive experience. The speed of scripts and processes is significantly improved and any waiting time is now acceptable. £350 well spent.
So what did I do:
| Start | Change | |
| Operating System | Windows 10 Home | No change |
| Processor | AMD A10-9700 Radeon R7 4 Core 4 Threads | AMD Rysen 7 5700G 8 Core 16 Threads |
| Motherboard | ASUS Prime A320M-K(AM4) | No change |
| RAM | 8 Gb | 32 GB |
| C Drive | SSD 256 GB | No change |
| D Drive | HDD 1TB | No change |
| H Drive | SSD 1 TB | |
| Swap files | 1 on C drive | 8 on H drive |
| Benchmark score | 2433 | 17202 |
| Benchmark time | 193.35 secs | 27.35 secs |
Before I replaced any kit I looked into what you can do with swap files and increased the number of swap files to 4. This improved my score from 2433 to 2782 and reduced the benchmark time from 193 seconds to 169 seconds
The final improvement was a x7 score in the benchmark score and a benchmark processing time of 1/7 of that achieved before any improvements undertaken.
Night Sky over Middleton 