|
| |
|
CBM Disk Transfer Benchmark V1.0 [2020] |
AKA :
DTB
Website :
https://www.obliterator918.com/dtb/
Credits :
Download :
Look for downloads on external sites:
Pokefinder.org
User Comment
Submitted by obliterator918 on 1 October 2020
| So a 281x score sounds right. I can't wait for the firmware update. :-D |
User Comment
Submitted by GideonZ on 1 October 2020
It is clearly slower than the REU. The bytes per second reading is, I think, correct. 112390 Bps in save, I got on my latest screenshot. :-)
Okay, totally understood. The TOD pin is indeed divided by 5 or 6 first. My apologies for suggesting something that is not easily feasible. You could use a timer A/B tandem without interrupts, but indeed, it is not too certain that custom load/save routines don't use them. |
User Comment
Submitted by obliterator918 on 1 October 2020
Wish I could be more precise, but the CIA TOD clock only gives you deciseconds. The only other option I can think of is to use the jiffy clock, which can be halted too easily. I didn't want to take the chance of a load routine redirecting the interrupt and the jiffy clock not getting incremented. I can't think of a way to get more precision than 1/10 of a second without interrupting the load process and affecting the outcome.
The DTB score is based on bytes per second. Is the printed Bps an unexpected one? Given that an REU can blit all of RAM in less than a second, a DMA load in .4 doesn't seem surprising... +/- a decisecond. :-) |
User Comment
Submitted by GideonZ on 1 October 2020
| Great little tool! I would suggest to use centi-seconds instead of deci-seconds. While testing the improved SoftwareIEC on the U64, I am getting a reading of 00:00:00.4 on load and 00:00:00.5 on save, which doesn't give me an accurate benchmark. (>200x) |
|
|
|
 | Search CSDb |
|
| Navigate | |
|
| Detailed Info | |
|
| Fun Stuff | |
· Goofs
· Hidden Parts
· Trivia
|
|
| Forum | |
|
| Support CSDb | |
|
 |  |
|
