spike: Test bigger framebuffer

[sbernauer]

Works in general, but for whatever reason

time:   [31.504 ms 31.558 ms 31.611 ms]
change: [+25.068% +25.395% +25.712%] (p = 0.00 < 0.05)
Performance has regressed.

Was also able to reproduce this looking at traffic levels and pixelflut client
CC @fabi321

[fabi321]

If I run the server with this change, it is slightly faster (15Gbit/s compared to 16Gbit/s), but the bench is within noise level for me. I'm not sure, whether the real world improvement is due to broken vnc server support.

[sbernauer]

Absolute agreed!
Would you be fine with closing this?

[fabi321]

sure, go ahead.

[sbernauer]

Alright, was a good idea I would say but yeah