Good discussion going on here. I’m putting a new build together at the moment about to go into production.
My daily setup is running 2x instances.
1x trading with 5 chartbooks, each has 7 charts. Only ever 10 charts displaying at a time.
1x global view market monitor. About 13 charts showing.
I have open GL turned on and make sure to only load the minimum required days with optimised ms loading on charts. Low on ladders, high on daily bar charts etc.
Where I am looking out an upgrade in order is
- Fastest possible time and sales updating.
- Lowest latency ladders.
- Faster chart loading.
- Faster initial load.
While 4 is not that important, 1 and 2 are extremely important.
I have found best performance for 1 and 2 was minimising no. Of chartbooks open. I think I have a medium to Conservative amount.
As I’m currently running an i7 7700k hyper threaded on standard ssds, I feel that the 4(8 virtual) cores are not enough as I reach about 70% to 80% cpu on market open. If I throw another 2 or 4 chartbooks in there, I max out and glitch.
So new system plan is
Cpu I9 10980xe
Ram 64gb 3600
2x2tb nvme ssd
Psu 1000w
GPU - NOT SURE. Thinking of recycling my gtm 1070ti 8gb. Or I could upgrade to the PNY4000 Cuda core.
So my question is, does anyone use a PNY for Sierra charts/ trading?
My monitor setup is
1x 55" samsung 4K UHD TV
2x 32" samsung 4h UHD
1X samsung 42" Wide-screen
So, which GPU would people think of using?
Many thanks. Great discussion above.
Best
Jester