Spider

No, it is normally reliable. Are you 100% you entered the correct code ? This might sound odd but try it again, if it persists, when on the code entry page itself being asked for a code, what is the byte value at 24965 ? It should be 2 The instruction itself is at 24964... 24964 JR NZ , 24968 ; jump if the code is wrong to ask again (or crash if its the second wrong attempt) If 24965 = 0 it reads like this 24964: JR NZ, 24966 ; jump if the code is wrong to accept it anyway: 24966: POP HL ; throw away the current return address from the stack 24967 RET ; Now the next one one will take us to the Title screen Enter something other than the correct code aka 1246 or 1111 etc Random: I think the codesheet for the AMS version is different (colouring aside) but I've not actually compared them, never saw the need.

Here. 😉 Comparison. https://spectrumcomputing.co.uk/forums/viewtopic.php?p=90487 Back on topic, as said please feel free to ask here, we can only try to help. 🙂

There are a couple of other disassembles too, with various levels of comments. Can I ask which platform you were considering ? It has been ported to quite a few (MM too) , just curious. 🙂 However (unlikely) we are 'missing' a Sam Coupe JSW and both MM and JSW for the Amstrad PCW256/512*** (the greenscreen wordprocessor) *** Don't laugh, there's some interesting titles on that machine, including Head Over Heels.

Thank you for the updated descriptor. 🙂 Always nice to have these up-to-date. I see the topic here now has decided*** to update itself to reflect that. *** Ongoing concern about it either updating immediately , a day later or in some cases never.

Great stuff! 🙂

Yes 🙂 That was partly the idea too. But I want to leave certain places in the code empty (3 bytes) where someone can easily CALL their own routine upon certain conditions in the game.

Made a couple of adjustments for now to see how it goes. I do see one small issue still.

I think the 'standard' MM and JSW run a bit too slow (honest thought) as with emulation I used to run them a bit faster. I do agree the third party edited ones are a lot harder and take into account (perhaps subconsciously) the erm "pace" ? I've seen myself with my own attempt at it, how fast MM and JSW can run with just a simple modification to the main loop , aka no actual detailed optimisation of the code in itself and also the next gain -if- the 'game buffer' was moved out of contended memory (easy with some emulators) I suppose when new on a real Speccy "back then" they seemed fast enough but these days we are spoiled a bit I suppose. 😕 I'm not sure. 🙂

Small update to my post with file, detailed in the description. I've not uploaded the txt notes on how as I'm not happy with the way I've written (and re-written) them yet.

Ah yes I saw this on Spectrum Computing the other day, its AGD based if I recall. I did have a quick go at the first two screens. 🙂

I -think- I provided those pokes 'as is' in both SP and BB versions as a "quickfix" to a request on Spectrum Computing, without thinking too much about how to make it better. Having said that I can also recall one of Andrew Broad's games (At least one) having the item collect sound added too. EDIT... Ah, yes it was here > https://spectrumcomputing.co.uk/forums/viewtopic.php?p=115680#p115680 Considered adding it to the 'slightly faster' MM version but as that was only to 'demo' the speed increase I did not really want to do anything other than bugfix with that.

I see the topic here was updated to include the 'new' version too. Thank you.

Yes (as I write this) that's correct. 🙂 I'm slightly pushed towards having them as one submission rather than two however I'll reply to the submitters PM in a moment regarding this, along with some other information too. The first .rzx was not published , the second one was in the topic as you've seen. I did not play to my best it was more a case of I wanted to explore and share. I can attempt the other version (the DS one) over the weekend I should think.

Was on my 'todo' list as well which still grows , but its one to remove now. Signup = Will investigate in a few days when I am able to. I -think- I can guess the likely causes.

Hi. Thank you again for the game! 🙂 The AMS and BBC levels were appreciated, the last AMS one being quite hard and I'm thankful that in BBC Meteor Storm the 'gates' were not electric. Oddly enough, I have (before I saw this message) just had a playthough again , this time (as I should of done first time) with your standard game as provided , so this one I -can- attach here: I did not rush hugely so its not a massively efficient way and I merely looked at what was going on in the cavern, then took a 'best guess' , I did lose one life probably shown in the playback and had about 4 rollbacks in total. Lost_Levels_Spider_1st_May_23.rzx Recorded in Spectaculator, should play back in that and Spin/Fuse/Zero.

I think that's it really , in that it does not matter how far MW falls if he hits one of these, he's safe (and bounces back up a bit) very similar to Dynamite Dan This just sort of makes him fall a lot faster , I can't recall -if- there's anything else to prevent a fatality with the fall as I do remember one cavern had this near the top, so he'd fall say 2/3rds the way down. Sorry the above is a bit vague.

Have just played it through, excellent as I found it not difficult for me. I can't put the .rzx here*** but I can do a video if needed. *** Only reason is I'd premodified the 'core' a little bit as an experiment before playing. Does -not- effect anything you did afaik.

@The-Red-Bayron Thank you for the submission and I've approved it. Files automatically generate their own topic to go along with the file itself (its linked to it) 🙂 https://jswmm.co.uk/files/file/172-manic-miner-the-lost-levels/ https://jswmm.co.uk/topic/773-file-manic-miner-the-lost-levels/

I'm similar but reversed 😉 I use Spin 99% of the time as I like the way it works plus the ease of 'small code edits' directly. To record .rzx's I use Spectaculator as Spin likes to pause it on each save (hugely annoying) plus I find it easier with the INS/DEL of Spectaculator. I do have Zero and Fuse installed but they are (probably) out of date and I only use them if I have a specific need. In particular Zero has an absolutely fantastic debugger, you can break on a memory location being read -or- (or not and) being written or when it contains a specific value and only that value. Also you can break on registers containing specific values (say "Break if HL is set to 34567) etc. It is not hugely friendly I found as the 'continue/start' ignores the settings you have to use the actual debug window to restart but in the past I have found it hugely helpful for backtracing, where for instance someone has provided a poke for infinite lives or such but then I want to backtrace to where they are actually initially set instead. Does save time looking 😉 🙂 I do remember ZX32 for Windows 9x and also X128. The latter 'full version' (or was it the former?) was released free I think eventually. I do recall one of these, the demo trial version after 10 minutes would not expire but set it to full speed.

I've not played it for a while, if I recall: The trampoline effect is similar to Dynamite Dan What I called the 'vaccum descent' , its sort of a fast downward drop compared to normal speed when you're over a certain floor tile. I think this 'feature' could be ignored possibly without detriment.

Would it help to know that usually I have ZXSpin set at either 105% or 115-120% ? As its (better) at slight speed increases compared to Specatulator. With the increase in speed here I felt it was fine at 100% 🙂 I honestly don't think its too fast*** *** Running it without contention (rewrite needed) and it was a bit "fun" 😮 🙂

There are videos here of those caverns too. The 'full' game itself is denied (afaik) however there are a few variants, some don't have the extra caverns and some do, plus the freebie 'demo' one that only contains a couple of caverns. The actual 'layout' is very similar in the code itself.

Yes no issues. The initial problems I thought about were the extra features aka trampolines or like "vacuum decent" but it would be possible to work around those.

Thank you. That's impressive. 🙂 I did look closely at the other two level sets for the Sam Coupe with a thought to porting those over, as in either one "40 cavern" game or perhaps two separate ones but it is still very much on the back-burner.

The above 'test file' ( while I learned, thank you N.S! 🙂 ) has the following changes (excluding cosmetics like loading picture and slightly altered piano on title screen) The main change is to the four main copy routines which originally used LDIR in the main loop are now replaced with a routine using a number of LDI's resulting in an increase in speed. These copy routines being: Copy master attributes to attribute buffer , 512 bytes Copy master screen pixels to screen buffer , 4096 bytes Copy pixels from screen buffer to real screen , 4096 bytes Copy attributes from attribute buffer to real screen , 512 bytes The relative increase in speed is quite noticeable. 🙂 The only other changes were three small insignificant ones: Willy faces the correct way upon entry on each cavern The highscore is preset to 010000 instead of 000000 Amoebatron's Revenge conveyor is alive. These are something I'd automatically do anyway, probably. EDIT... Update: New loading screen (as logo then full) , Andy Green 2018 Item collection sound (not in exactly the same place as other games) Score status line flicker reduced (not eliminated but reduced) Demo mode now plays the full ingame tune per cavern instead of half of it Pressing BREAK during gameplay will update the highscore if its new MM_SP_Faster V2.tap No other changes, plenty I want to do but it is a demo/basic building block for others.