From what I understand SH3 games have a little intentional lag built in, and I absolutely agree there is a small amount of additional lag with mame. I do question how many people would notice if their pcb somehow had the same level of lag when they woke up one day though. I own a ton of original software I purchased new from retailers, but mame on a pro CRT is so gorgeously convenient it's simply too hard for me to ignore personally.
I could be very wrong here, but from memory I believe Trap confirmed the lag in the SH3 titles somewhere on these boards a while ago. While it's not an SH3 title, he did a great job of making an arrange mode for Ketsui, which is more than enough to let me know he knows his stuff.
Personally I want to know how much he would charge me to change the color of the bullets in futari
finisherr wrote:I've bought just about all of the shooters I'd want to play that are compatible with an American 360. The stuff I will probably emulate includes DOJ, DDP, Ketsui, and Mushihimesama. If there were region free releases of these games, I'd have no problem buying them. But you know...there aren't.
If you are considering that path, why not hack your Xbox to play Japanese games and buy them?
A ban from Xbox Live Gold, among other things. Never mind the fact that some people don't have the knowledge or resources needed to hack a Xbox 360.
I'm really not willing to potentially destroy my XBox. If I had an extra, well, then I'd have something to work with.
Compared Espgaluda II in the new Mame to X360 and I hated it. Besides Zeshikai mode being completely messed because of blitter settings it just controlled terribly with at least 3 frames of additional lag plus the sound was garbage. All of the non-Shmupmame SH3 stuff is no fun to play in Mame if you are used to PCB or console versions. It may depend on your sensitivity to these things and how serious you are about playing your best but I wouldn't bother with such great games released officially and easily purchased for a good price.
kathy wrote:From what I understand SH3 games have a little intentional lag built in
I've read about that a few times but was it actually confirmed or documented somewhere ?
Wouldn't be a surprise though.
One indication is the Frame Delay setting in the Mushi Futari port which are as follows:
0V(Tearing), 1V(Fast), 2V(PCB), 3V(Slow)
This would indicate that the PCB possibly has 2 frames of delay put in. ZOMG!
I wouldn't assume the lag is "put in" so much as forced through a limitation of hardware. Would it make sense as a developer to say "yeah, controls seem good... but it needs more lag." What I wonder is what does the "V" stand for. I would assume it means frames but in principle I don't think there is such thing as zero lag as your input may not register until the next frame is drawn.
casualcoder wrote:Compared Espgaluda II in the new Mame to X360 and I hated it. Besides Zeshikai mode being completely messed because of blitter settings it just controlled terribly with at least 3 frames of additional lag plus the sound was garbage. All of the non-Shmupmame SH3 stuff is no fun to play in Mame if you are used to PCB or console versions. It may depend on your sensitivity to these things and how serious you are about playing your best but I wouldn't bother with such great games released officially and easily purchased for a good price.
Are you sure your emulation is running at 100% speed (Press F11). I had no sound issue with that game, but you need at least a 3Ghz C2D to run the emulation at full speed. Ibara needs a 3.33Ghz C2D. Also are you turning on the blitter delay in the menu? Tab> Game Settings> Blitter Delay On?
The blitter never felt right to me but even if I tweaked it the lag was a big turn-off. I compared the final boss pattern and the lag made my usual movements stupid hard to produce consistently. Not worth it for me. I was hoping to use it for savestate practise but no dice.
casualcoder wrote: I wouldn't assume the lag is "put in" so much as forced through a limitation of hardware. Would it make sense as a developer to say "yeah, controls seem good... but it needs more lag." What I wonder is what does the "V" stand for. I would assume it means frames but in principle I don't think there is such thing as zero lag as your input may not register until the next frame is drawn.
That's a fair call and makes total sense.
Personally I can't get Ibara running right and I haven't had much to do with Galuda II. I own the port but much prefer playing Futari and some of the previous gen titles.
The inclusion of SH3 code in mame might be a step forward in getting the sub par games working better though, which is great news IMHO.
this will never work correctly with the blitter rate settings like this. trap15 mentioned this several times that this is a very dirty workaround for emulating slowdown and that he was able to produce a much more sophisticated code with 100% accurate slowdown.
There's input lag in SH3 titles on hardware ? I never noticed it and I'm generally pretty sensitive to it.
On the other hand the Raizing trinity have so much lag that unless one of them is the only game I'm playing they make me want to throw the cabinet out of a window, it's revolting. If someone could fix Garegga/Batrider input lag on real hardware I'd be happy to pay them.
matrigs wrote:this will never work correctly with the blitter rate settings like this. trap15 mentioned this several times that this is a very dirty workaround for emulating slowdown and that he was able to produce a much more sophisticated code with 100% accurate slowdown.
trap15, if this is true and not a misunderstanding, what is keeping you from submitting that code to the MAME project?
Contempt. Also, they're ridiculously simple changes, so if they want to fix it, it's not hard for them to actually do it. I'll say it again: the key is not in the GPU delay, it's in the CPU. Read the datasheet.
Great news. Time to compile again my build. Is Akai Katana added? Have they improved the emulation of Pink Sweets and Muchi Muchi Pork? As regard what has been discussed, I must say I enjoy Cave games a lot in MAME. I play them in a CRT monitor in native low res. The only fact that I can kill all slowdowns makes them better than in the original hardware, so I can forgive any minor issues in emulation.
trap15 wrote:Contempt. Also, they're ridiculously simple changes, so if they want to fix it, it's not hard for them to actually do it. I'll say it again: the key is not in the GPU delay, it's in the CPU. Read the datasheet.
Would you mind posting your .diff file so that people could compile their own MAME versions and enjoy those games with proper slowdown?
cools wrote:
On the other hand the Raizing trinity have so much lag that unless one of them is the only game I'm playing they make me want to throw the cabinet out of a window, it's revolting. If someone could fix Garegga/Batrider input lag on real hardware I'd be happy to pay them.
cools wrote:
On the other hand the Raizing trinity have so much lag that unless one of them is the only game I'm playing they make me want to throw the cabinet out of a window, it's revolting. If someone could fix Garegga/Batrider input lag on real hardware I'd be happy to pay them.
That would be a dream come true.
The only sensible response to this ridiculous notion is lrn2play.
casualcoder wrote: I wouldn't assume the lag is "put in" so much as forced through a limitation of hardware. Would it make sense as a developer to say "yeah, controls seem good... but it needs more lag." What I wonder is what does the "V" stand for. I would assume it means frames but in principle I don't think there is such thing as zero lag as your input may not register until the next frame is drawn.
Well yeah, put in is probably wrong, I agree. The only thing I can think of the V could stand for is frames of V-sync.
trap15 wrote:Contempt. Also, they're ridiculously simple changes, so if they want to fix it, it's not hard for them to actually do it. I'll say it again: the key is not in the GPU delay, it's in the CPU. Read the datasheet.
Would you mind posting your .diff file so that people could compile their own MAME versions and enjoy those games with proper slowdown?
Nope. It's really hacky and awful because I didn't care too much, which is another reason I won't release them. I doubt in the ability of MAMEdev to make a better implementation, but it'd be fun to see how awful they do it, if they ever even do.
trap15 wrote:Contempt. Also, they're ridiculously simple changes, so if they want to fix it, it's not hard for them to actually do it. I'll say it again: the key is not in the GPU delay, it's in the CPU. Read the datasheet.
why you so sure, if the cause of slowdons is CPU why it usually used only about 30% ?
I've readed SuperH datasheets million times, where exactly to look at ?
if you have some ideas or know something - feel free to share it here or on #messdev or smth
| My games - http://www.emphatic.se
