Thomas Jentzsch Posted April 6, 2010 Share Posted April 6, 2010 (edited) I knew your Clonespy was producing these percentages, but I was unsure if it'd pick up small packets of code bring moved around. [/Quote] Yes and no. CloneSpy works like a pseudo-compressor. It searches for identical code sequences and "stores" them with some overhead. Short sequences create a lot of overhead (too short sequences are skipped). So with e.g. 2 bytes (not sure, have to check the code) overhead/sequence, a 4 byte table would be stored with 50% space gain. Longer sequences have less overhead and create more space gain. HOM3 does a simply binary diff, so unlike CloneSpy it won't find moved around code. Edited April 6, 2010 by Thomas Jentzsch Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted April 6, 2010 Share Posted April 6, 2010 I knew your Clonespy was producing these percentages, but I was unsure if it'd pick up small packets of code bring moved around. [/Quote] Yes and no. CloneSpy works like a pseudo-compressor. It searches for identical code sequences and "stores" them with some overhead. Short sequences create a lot of overhead (too short sequences are skipped). So with e.g. 2 bytes (not sure, have to check the code) overhead/sequence, a 4 byte table would be stored with 50% space gain. Longer sequences have less overhead and create more space gain. BTW: How does HOM3 work here? HOM3 does a direct line by line comparision. Picking up a byte from towards the end and shifting it to the top will give you thousands of line differences. In truth I would only count that as just 1 difference as I don't consider shifted addresses as differences, as long as they go to the same place in the routine. I would like to know the limit before a sequence is ignored, if you can dig that up Thomas. It'd be good to know. Quote Link to comment Share on other sites More sharing options...
Thomas Jentzsch Posted April 6, 2010 Share Posted April 6, 2010 I would like to know the limit before a sequence is ignored, if you can dig that up Thomas. It'd be good to know. The basic overhead is two. There are some additional tweaks in CloneSpy e.g. identical offsets inside a sequence don't break the sequence, just increase overhead. The used compression algorithm is LZSS, which is pretty good. But there are better ones. BTW: xdelta is a tool to create patch files whichs convert file A into file B. The smaller the resulting patch file, the more identical the two files are. From the documentation it seems like it uses a very similar approach like Clonespy. To patch Air Raid into: Space Jockey (U.S. Games Corporation) (NTSC): 1275 bytes Space Jockey (Carrere Video) (PAL): 1392 bytes Space Jockey (Unknown) (PAL): 1540 bytes UFO (32 in 1) (Atari) (PAL): 1576 bytes Space Jockey (208 in 1) (Unknown) (PAL): 1551 bytes (Note: You have to remove the overhead from the patch files 7 bytes + length of both filenames. E.g. 1275 -> 1141 bytes) Quote Link to comment Share on other sites More sharing options...
Mirage Posted April 6, 2010 Share Posted April 6, 2010 EPROMS NOT GLUED DATED 1975 SHOW THAT 10 OUT OF 9 PEOPLE DESPITE VIOLENCE AGREE THAT AIR RAID CARTRIDGE IS PROVEN BEYOND DOUBT TO NOT HAVE BEEN REPRODUCED LATER THAN THAT DATE 10 out of 9 people seems a little hard to do. He was counting one of the voices in his head. Quote Link to comment Share on other sites More sharing options...
A.J. Franzman Posted April 6, 2010 Share Posted April 6, 2010 EPROMS NOT GLUED Gluing of EPROMS is not a standard practice. What are you talking about? 10 OUT OF 9 PEOPLE DESPITE VIOLENCE AGREE 10 out of 9 people agree? Did you conduct and compile this survey yourself? EPROMS [...] DATED 1975 SHOW [...] THAT AIR RAID CARTRIDGE IS PROVEN BEYOND DOUBT TO NOT HAVE BEEN REPRODUCED LATER THAN THAT DATE Firstly, proving that something like that never happened is a logical impossibility. Secondly, assuming that the IC markings are legitimate, this would only prove that the cartridge was made in or later than 1975 (not how much later... it could have been many years after). Thirdly, I doubt that the Air Raid cartridge has ever been "reproduced". Based on what we know about the making of custom plastic items, and the rarity of the game, it seems that what it cost to originally produce it (did you mean "produced" instead of "reproduced"?) probably far exceeded the profit generated from selling it. It's possible that the company and/or person responsible for the game was bankrupted by it. Finally, what drugs are you on? That's gotta be some good $#!+. Quote Link to comment Share on other sites More sharing options...
CPUWIZ Posted April 6, 2010 Share Posted April 6, 2010 PM sent. Quote Link to comment Share on other sites More sharing options...
hhwolfman Posted April 6, 2010 Share Posted April 6, 2010 PM sent. :rolling: :rolling: Quote Link to comment Share on other sites More sharing options...
Dino Posted April 6, 2010 Share Posted April 6, 2010 EPROMS NOT GLUED Gluing of EPROMS is not a standard practice. What are you talking about? 10 OUT OF 9 PEOPLE DESPITE VIOLENCE AGREE 10 out of 9 people agree? Did you conduct and compile this survey yourself? EPROMS [...] DATED 1975 SHOW [...] THAT AIR RAID CARTRIDGE IS PROVEN BEYOND DOUBT TO NOT HAVE BEEN REPRODUCED LATER THAN THAT DATE Firstly, proving that something like that never happened is a logical impossibility. Secondly, assuming that the IC markings are legitimate, this would only prove that the cartridge was made in or later than 1975 (not how much later... it could have been many years after). Thirdly, I doubt that the Air Raid cartridge has ever been "reproduced". Based on what we know about the making of custom plastic items, and the rarity of the game, it seems that what it cost to originally produce it (did you mean "produced" instead of "reproduced"?) probably far exceeded the profit generated from selling it. It's possible that the company and/or person responsible for the game was bankrupted by it. Finally, what drugs are you on? That's gotta be some good $#!+. Quote Link to comment Share on other sites More sharing options...
Omegamatrix Posted April 7, 2010 Share Posted April 7, 2010 The basic overhead is two. There are some additional tweaks in CloneSpy e.g. identical offsets inside a sequence don't break the sequence, just increase overhead. The used compression algorithm is LZSS, which is pretty good. But there are better ones. That is actually a lot finer than I thought it'd be. The Clonespy percentage should be very close then indeed. Quote Link to comment Share on other sites More sharing options...
+5-11under Posted April 7, 2010 Share Posted April 7, 2010 I would like to know the limit before a sequence is ignored, if you can dig that up Thomas. It'd be good to know. The basic overhead is two. There are some additional tweaks in CloneSpy e.g. identical offsets inside a sequence don't break the sequence, just increase overhead. The used compression algorithm is LZSS, which is pretty good. But there are better ones. BTW: xdelta is a tool to create patch files whichs convert file A into file B. The smaller the resulting patch file, the more identical the two files are. From the documentation it seems like it uses a very similar approach like Clonespy. To patch Air Raid into: Space Jockey (U.S. Games Corporation) (NTSC): 1275 bytes Space Jockey (Carrere Video) (PAL): 1392 bytes Space Jockey (Unknown) (PAL): 1540 bytes UFO (32 in 1) (Atari) (PAL): 1576 bytes Space Jockey (208 in 1) (Unknown) (PAL): 1551 bytes (Note: You have to remove the overhead from the patch files 7 bytes + length of both filenames. E.g. 1275 -> 1141 bytes) I'm curious what kind of result you get when comparing to a completely different game (or is that the UFO game that is mentioned)? Thanks, 5-11under Quote Link to comment Share on other sites More sharing options...
Zap! Posted April 7, 2010 Share Posted April 7, 2010 PM sent. Yup, here it comes! He accepts your offer and the auction is pulled! Quote Link to comment Share on other sites More sharing options...
Tanman Posted April 7, 2010 Author Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding Quote Link to comment Share on other sites More sharing options...
Dino Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds Quote Link to comment Share on other sites More sharing options...
Cebus Capucinis Posted April 7, 2010 Share Posted April 7, 2010 I resent that. I am a hip swinging lounge lizard! Quote Link to comment Share on other sites More sharing options...
Dino Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds BTW, nice going on your auction so far.... very nice! Quote Link to comment Share on other sites More sharing options...
+Nathan Strum Posted April 7, 2010 Share Posted April 7, 2010 The CloneSpy values are percentages, so 56% of code is shared (including repeated code e.g. in the "new" 2k bank). So does that mean my boxed Space Jockey is worth 56% of what Air Raid is? 5.6%? .56%? .056%? .0056%? Quote Link to comment Share on other sites More sharing options...
Zap! Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds I'm a geek not a nerd! Quote Link to comment Share on other sites More sharing options...
Dino Posted April 7, 2010 Share Posted April 7, 2010 The CloneSpy values are percentages, so 56% of code is shared (including repeated code e.g. in the "new" 2k bank). So does that mean my boxed Space Jockey is worth 56% of what Air Raid is? 5.6%? .56%? .056%? .0056%? .56% is still $70 Quote Link to comment Share on other sites More sharing options...
Dino Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds I'm a geek not a nerd! I'm both, and proud of it! Quote Link to comment Share on other sites More sharing options...
gamme Posted April 7, 2010 Share Posted April 7, 2010 Another case of the April Fool joke kept inside, good luck all. Quote Link to comment Share on other sites More sharing options...
atwwong Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds I'm a geek not a nerd! I'm both, and proud of it! It's the in thing, nowadays... Quote Link to comment Share on other sites More sharing options...
pangasinan Posted April 7, 2010 Share Posted April 7, 2010 This thread has been highjacked by nerds!!! hehe just kidding This whole WEBSITE is full of nerds I'm a geek not a nerd! I'm both, and proud of it! He said geek not greek Quote Link to comment Share on other sites More sharing options...
Thomas Jentzsch Posted April 7, 2010 Share Posted April 7, 2010 I'm curious what kind of result you get when comparing to a completely different game...? If you e.g. compare Pitfall! with Pac-Man you get ~3600 bytes. So they share maybe 10% code. Which is quite normal for unrelated games. Quote Link to comment Share on other sites More sharing options...
GeneralHellfire Posted April 7, 2010 Share Posted April 7, 2010 I'm out. I was interested, but not $10K+ interested.... or in other words, I'm crazy, but not that crazy. I'll just keep my reproduction cart and hope for a Repro box at some point. Even better, if someone makes a reproduction cart (t-handle) with Repro box. With all this hype, I bet this would make a good limited production run. Heck, they might even sell more than Men-a-vision did with the original Quote Link to comment Share on other sites More sharing options...
Shawn Posted April 8, 2010 Share Posted April 8, 2010 I'm out. I was interested, but not $10K+ interested.... or in other words, I'm crazy, but not that crazy. I'll just keep my reproduction cart and hope for a Repro box at some point. Even better, if someone makes a reproduction cart (t-handle) with Repro box. With all this hype, I bet this would make a good limited production run. Heck, they might even sell more than Men-a-vision did with the original I have some Brazil TRON t-handle carts in my watch list right now contimplating doing just that, a few CIB baby blue t-handle Air Raid repros. I'm just worried if I sell them outside of the well known collecting circles they would be abused\misrepresented for resale. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.