WHDLoad MantisBT - RuffNTumble |
| View Issue Details |
|
| ID | Project | Category | View Status | Date Submitted | Last Update |
| 0004174 | RuffNTumble | [All Projects] General | public | 2019-07-30 22:18 | 2022-04-14 23:55 |
|
| Reporter | sanjyuubi | |
|---|
| Assigned To | JOTD | |
|---|
| Priority | normal | Severity | minor | Reproducibility | always |
|---|
| Status | closed | Resolution | unable to reproduce | |
|---|
| Platform | | OS | | OS Version | |
|---|
| Product Version | | |
|---|
| Target Version | | Fixed in Version | | |
|---|
| Machine | A500 |
|---|
| CPU | 68EC020 |
|---|
| CPUSpeed | 25 |
|---|
| ChipSet | OCS |
|---|
| GFXCard | None |
|---|
| ChipMem | 1 MB |
|---|
| FastMem | 9.5 MB |
|---|
| Workbench | OS 3.1 |
|---|
| KickROM | 40 - Kick 3.1 |
|---|
| KickSoft | None |
|---|
| WHDLoad | 18.5 |
|---|
| imported | yes |
|---|
|
| Summary | 0004174: Hello! I've been wondering if to commit this issue on Mantis Bugtracker or not, |
|---|
| Description | GameVersion: english,pal,2 disks
SlaveVersion: 2.3
Hello!
I've been wondering if to commit this issue on Mantis Bugtracker or not, so I decided to present my observation first. While playing around with 68020 accelerator, I noticed something is not entirely alright with fluency while watching demo, the issue is noticeable especially on the bridge with enemies shooting spread bomb projectiles from beneath in 2nd world (1st demo played by the game), the game is stuttering slightly in tight moments. The game launched from floppy doesn't have this problem and is playing fluently. I recorded those demos to compare their length and peculiar stuff comes to the light. I had three accelerator setups, one 68020 clocked at 25MHz, one clocked 28MHz and one clocked at 28MHz with memory working at 1 waitstate (performance result in Sysinfo is slightly below 25MHz with 0ws). While floppy version doesn't lag at all on all setups, WHDLoad one lags about 4s behind at the end of the demonstration on 25MHz accelerator, both 28MHz setups have about 2s lag with no discern between 0ws and 1ws. Floppy version on 7MHz is only a fraction of second behind accelerated game, you may think that WHDLoad introduces some overhead, but it's debunked by the most baffling result is WHDLoad gameplay at 7MHz lagging less than accelerated games.
Here's the link of the video for comparision:
https://www.youtube.com/watch?v=zMsLkgqIeYk&feature=youtu.be
Is this something worth investigating? Technically, it's playable. |
|---|
| Tags | No tags attached. |
|---|
| Relationships | |
| Attached Files |  RuffNTumble.zip (2,561) 2021-04-26 21:41
http://mantis.whdload.de/file_download.php?file_id=1202&type=bug |
|---|
|
Notes |
|
|
(0006934)
|
|
JOTD
|
|
2019-07-31 23:47
|
|
|
7Hz is all right? you mean 68000? then you can try NOVBRMOVE tooltype on other machines and report here. |
|
|
|
|
7MHz at 68000 seems almost alright, if you'll fast forward to the end of the movie you will notice that demo on whdload plays about 0.5-1s longer than floppy version on 68000.
NoVBRMove didn't do anything in regards of fluency, i tried also noautovector, expchip, cache with no avail.
I'll try look into the matter with WinUAE and check if such behavior happens there as well. |
|
|
|
(0009897)
|
|
JOTD
|
|
2021-04-08 22:33
|
|
then try with NOCACHE on the 68020. I've noted that some games can be super-slow if "fastest possible" config is enabled in winuae
for instance Red Zone, or Turrican startup... |
|
|
|
|
Hi!
I checked Ruff in WinUAE and exactly same thing is happening, Ruff ran from floppy on 68020 is running faster than with 68000, but under whdload it is running slower on 68020. NOCACHE have no impact.
https://www.youtube.com/watch?v=2lHgRXcdGxE
Should i provide winuae with configuration files? I'm ClassicWB, KS3.1, 8MB fast, 1.5MB slow and cycle-exact settings, 68020 has been set to 28MHz. |
|
|
|
(0010109)
|
|
JOTD
|
2021-04-26 21:41
(Last edited: 2021-04-26 21:44) |
|
can you test the attached slave?
This can't be a specific issue, but rather a whdload issue @Wepl any idea?
|
|
|
|
|
Tried this slave and nothing really changed, demo had 3s lag behind "floppy run" on 68EC020, but when i switched to 68000@28MHz in Winuae, the demo had only 1s lag.
I also tried "fastest possible" and while in case of 68000@28MHz it didn't change a thing compared to cycle exact (1s lag), with 68EC020 it lagged even more (7s total compared to game launched from floppy with 68EC020 - cycle exact/fastest possible runs same speed from floppy).
This is weird. |
|