New Game Development

Ran beautifully off the USB though.

What format was the USB stick. I assume it was FAT. Were you using Fat32fs?
This has a very aggressive read ahead / write behind buffer system (512KB for each).

Hi Jeff,

On the Raspberry Pi 1, I have always run on the SD card with no MP3 playback issues. Yes the USB was FAT32.

Looking at http://elinux.org/RPi_SD_cards, I suspect my MicroSD card choice was probably poor. I’ve had the occasional freeze in RISC OS on it too.

Tony

Just a quick update. Making good progress on POC for new games. The next released game is expected to use 24 bit colour and (hopefully) parallax scrolling. Two games are being developed in parallel both of which have arcade/puzzle elements.

The website is now up: http://www.amcog-games.co.uk.

Also the RISC OS Blog and Riscository have uploaded some more pages on Overlord, which is greatly appreciated:
http://www.riscosblog.co.uk/?p=599
http://www.riscosblog.co.uk/?page_id=448
http://www.riscository.com/2015/overlord-now-available-from-store/#more-3853

Best Regards,

Tony

Further updates.

I’ve uploaded a video of Overlord:
https://www.youtube.com/watch?v=NRjs32-ARS4

Also, Risc OS Blog have reviewed the game:
http://www.riscosblog.co.uk/?p=318

I’m also exhibiting at the Recursion fair in a couple of weeks.

I did a talk a ROUGOL this week which I really enjoyed doing. Took some soundings on a new ‘quest’ arcade-adventure game that I am writing. This is planned to be an isometric 3D game in true colour which is intended to employ 8 way scrolling. The most significant jobs that I have at the moment are completing the character graphics & puzzle design. Hopefully I’ll get some help with graphics from an artist that I know… Will see if I can get some pencil drawings done this weekend.

In terms of game play, the target time to complete this game, assuming that you know all of the solutions, would be around 2 hours (not sure how many puzzles that would involve). Just wondering, does this sound reasonable, too long or too short based on user expectations?

Thanks and Regards,

Tony

My 2 cents:

I suspect that 2 hours completion time, assuming that all solutions are known, might be overly long nowadays. I think most game players just want their quick gaming ‘fix’.

I hope my opinion doesn’t in any way discourage you, however!

David.
—

I suspect that 2 hours completion time, assuming that all solutions are known, might be overly long nowadays. I think most game players just want their quick gaming ‘fix’.

My two centimes to counter David’s two cents…

I think “gaming” falls into two camps. The first, the one David is referring two, is the people that want a quick fix. A bit of punching, shooting, or whatever, but nothing too challenging.

The second, the ones who want a story and to be absorbed into it. I’m looking at my collection of PS2 games – Project Zero / Fatal Frame (creepy as hell! – short review), Hitman, Grand Theft Auto 3 : Liberty City (great fun with the many cheats available – hehehe), Project Eden (awesome intro sequence), Ghost In The Shell – Stand Alone Complex (she's so cold and human/it's something humans do/she stays so golden solo/she's so number nine... – gameplay), Sims 2, Sims 2 – Castaway, Enter The Matrix (or something like that), Resident Evil – Code Veronica X (which I utterly suck at), and Tomb Raider Legend (I kinda suck at this too).

The only game I’ve seen in recent years that would convince me to buy a newer console if I had the money is an epic called The Last Of Us (trailer).

As you might guess, these take time. Sometimes a LOT of time. Here’s a survey I found of how long it took to complete The Last Of Us – http://www.giantbomb.com/forums/the-last-of-us-7917/how-long-did-it-take-you-to-finish-the-last-of-us-1443176/

What I would ask is please include cheats. I don’t tend to play GTA3 that much, but I do like to start it, give myself all the good cheats, and just go exploring the game world. Any decent game will have a whole lot of things to discover, it seems such a shame if these are wasted on the lamer player being continually slaughtered within the first few minutes. This is why I wander around exploring Liberty City, but barely even remember that I have Resident Evil. After all, if I suck at the gameplay (and growing up with a debugger instead of a joystick, I pretty much entirely suck at anything more advanced than Chucky Egg¹), it would be nice to at least appreciate the aesthetic.

¹ And boy was I annoyed after “breaking into” the computer room at school every evening and for an entire weekend in order to learn (taking notes of failures) and then complete Chucky Egg … and there was NO “whoo-hoo-you-rock” screen. It just started over from the beginning. BAH!

There’s such a thing as “too long” too: By the end of Dragon Quest on PS2 (~60 hours), it felt more like I was slogging through it than playing it.

Thanks for the comments David, Rick and Chris.

I plan to make the game save-able, so that you can just play it for 20 minutes and stop, then come back to it if you choose. Much like with Overlord, I want to try and give players choices and different ways of playing the game to suit them. However this game is intended to have more of a narrative, as it is based (to an extent) on an unpublished fantasy novel (~70’000 words) that I have written which I might bundle with the game as a PDF.

I am always happy to include cheats. Regarding aesthetic, I’m not going to employ photo-realism, but am using pencil drawings as a start point which I’m basing on a manga style for the characters. These are reduced substantially and computer coloured. The aspect of the game is planned to be isometric 3D in 24 bit true colour. Some elements are based on real physical things (trees for example) that have been processed and stylised.

The 2 hour figure was suggested during my presentation at ROUGOL and would refer to a perfect game rather than exploratory play. Will have to see if I can achieve that however.

Chris:

There’s such a thing as “too long” too: By the end of Dragon Quest on PS2 (~60 hours), it felt more like I was slogging through it than playing it.

I prefer a longer game, but it does need to be very engaging if it takes that amount of time!

On the other hand, I found some of the Call of Duty games to be disappointingly short – including one where I struggled with one particular level because I did it incorrectly (but still eventually got through it). Those I can typically spend a weekend on and play from start to finish; taking into account real life, I’d guesstimate twelve to fifteen hours at most.

I reckon my ideal would be thirty to forty hours – but the game has to be engaging. (And/or it should allow free roaming and free play, like the GTA games, for when you just want to have some mindless fun).

However, I’m talking here about big budget (and therefore pricey) games. Cheap (and free) games are a different matter (as anyone who’s ever played any of mine would know – piece of cake, the lot of ’em!)

Anthony:

The 2 hour figure was suggested during my presentation at ROUGOL and would refer to a perfect game rather than exploratory play. Will have to see if I can achieve that however.

Quite. People have said to me at shows that they’re stuck in Quicksand, and don’t understand why I say it can be completed in less than an hour. It can (much less, probably) but when I’ve said that, I’ve also said /with a map and/or full knowledge of the puzzles and their solutions!/

There’s no save option (because of a silly approach I took to the script language in Trellis¹) but it’s short enough that provided you map it as you play, it shouldn’t matter if you have to start again.

¹ I’ve already decided on a fix for that – but until I deal with the redraw bug in the desktop version, I won’t bother implementing it.

60 hours is considered long these days? I remember playing Rage Racer on the PS1 for probably 60 days to win every race. The last level alone took perhaps 40 of these 60 days. You needed to drive pixel-perfect for 3 consecutive laps. RR7 on the PS3 was a piece of cake compared to this.

Thanks for the thoughts. Once I’ve been to Recursion I will be spending more time on getting this new game developed.

In the meantime, I’ve written an update to Overlord that will be available at Recursion and then will be free for all existing Overlord users. I understand that when I upload the new version to the !PlingStore existing users will receive a notification that a new version is available. If you bought a copy at a show, then the PlingStore code inside the box can be used to download the game.

If anyone has difficulties then please drop me a line. I will be updating the website with this information soon.

I’ve added a mouse sensitivity control to the start screen as some users found the mouse too sensitive.

A new power up that fires duel bullets (which is blue) has also been added.

In addition, I’ve updated the code so that the mission special objects are less likely to occur in quick succession. Which should improve the play-ability of the game.

Currently these changes are being tested.

Overlord V1.40

Overlord has been updated and now includes:

• A mouse sensitivity slider.
• A new (blue) ‘duel fire’ power-up.
• Improved distribution of mission objects.

This update is free for existing users and is available through the PlingStore. If you bought a CD at Wakefield or the ROUGOL presentation, then the PlingStore code in the CD box may be used to download version 1.40.

AMPlayer has a couple of small (maybe trivial) utilities in its build in C & BASIC that look they’ve been a victim of a cvs2svn conversion (ASClean and vutab). The svn property ‘svn:eol-style’ is set to native line endings

Indeed it did. Having looked at this

• ASClean wasn’t needed any more, since vaguely recent versions of objasm can preprocess the source for you so don’t need to go on a cleanup pass (removing #line directives)
• vutab had got mangled by LF→CRLF translation, so I just used !Edit to reverse the process and retokenised the BASIC
• Many many of the source files had the native EOL property set which meant they ended up with CRLF line endings if you checked them out on Windows and they then don’t assemble
• A lot of the binaries (eg. DrawFiles) weren’t marked as binary so got screwed too

Fortunately resident guru Theo was happy to help out reimporting the repository to sort out the line endings problem, then I just re-added the helper C programs. They seem to be 26 bit, so you need an emulator or Risc PC to run them.

All that waffle aside, I was then able to make a few minor updates to remove the unaligned accesses (only one in the assembler sources, plus the relevant switch for the C compiler). I’ve created a test version of !AMPlayer 1.40 for people to try.

My whole MP3 collection fits on about 3 floppies, and they all played fine on a BeagleBoard xM with alignment exceptions on, but please give it a run with your more extensive collections!

My whole MP3 collection fits on about 3 floppies, and they all played fine on a BeagleBoard xM with alignment exceptions on, but please give it a run with your more extensive collections!

???? One single MP3 shouldn’t fit across three floppies. Unless you’re talking iomega zip discs or something.

My phone tells me that my current MP3 collection is 3.28GiB and 530 tracks, so a quick bit of maths would suggest the simple average file size is 6.3MiB per song (though in reality it depends on the song length, bitrate/quality, and whether the song was encoded VBR or CBR).

???? One single MP3 shouldn’t fit across three floppies

Not everyone accepts relatively low quality of mp3 encoding. On good audio equipment there’s quite big difference between CD-Audio (or vinyl/SACD/192kHz lossless) and even highest quality lossy file (be it mp3, Spotify’s ogg or aac from Apple iTunes store).

Not everyone uses it as primary means of music access then :-)

http://archimago.blogspot.ca/2013/02/high-bitrate-mp3-internet-blind-test_3422.html

While a lot of mp3s suck, done well they can be virtually indistinguishable from lossless. It’s just a shame that the early days of mp3 were marked by 128kbit encoding and poor algorithms.

However note that you’ll need to create your “test” mp3 from your supposedly high quality copy, else what you may end up hearing are differences in mastering, not the format itself. This is especially notable with, for example CD vs Vinyl. There are great differences between the two, but if you burn your own CD recorded from your vinyl, then it can be extremely difficult to tell the difference. Unfortunately for us, the ‘00 decade was the decade of CD “loudness” where producers cared little for audio fidelity and just wanted their product to sound louder than the one that played before. I suppose if you are listening to hip hop then it probably doesn’t matter, but any music with actual fidelity to it (and I don’t just mean classical, symphonic rock – that you might expect to benefit from being loud, suffers) sounds poor.
Somewhere along the way it was decided to discard the range of sounds capable of CD audio and just crank it up until it starts clipping. Well, sorry, any sane person who wants the sound louder ought to just turn up the volume. If the source is wilfully damaged in order to lie to punters, it will sound horrible no matter how expensive the equipment. Gold plugs and speakers that cost the same as a family car won’t rescue a badly mastered original… So I wonder how many who prefer vinyl over CD (or SACD over CD) are comparing the attributes of the formats, or the attributes of the mastering?

BTW, 192kHz? You want your music to intentionally sound worse ? http://xiph.org/~xiphmont/demo/neil-young.html

Edited to fix predictive text fail and add some (!) detail.

Not everyone accepts relatively low quality of mp3 encoding. On good audio equipment there’s quite big difference between CD-Audio (or vinyl/SACD/192kHz lossless) and even highest quality lossy file (be it mp3, Spotify’s ogg or aac from Apple iTunes store).

I once took part in the “Hörtest” by German computer magazine c’t. You got 10 WAV files – one was the “original” and the other 9 were compressed with various codecs with bitrates between 128 and 192 kbit/s. You had to rank the files according to perceived quality. As a special surprise, there were only 8 codecs involved, the 9th file was actually the original.

Despite the low bitrate, it was really really hard to find differences between the uncompressed original and their compressed counterparts. In a later test, c’t found out that in a double blind situation, 256 kbit/s MP3 and uncompressed CD was not distinguishable at all.

If you need a lesson in humility, I would recommend doing such a test – just to find out how bad your expensive audio equipment really is. If you have a really good headphone (and really good ears!), you might stand a chance.

???? One single MP3 shouldn’t fit across three floppies

I took Sprows comment to mean ‘All the MP3 files he has total less than 3×1.6MB’ i.e. a small number of short MP3s!

Thanks Sprow. I’ll see if I can try this out with Legends of Magic. It would be great to have 1 MP3 solution rather than 2 in the code.

Regarding sound quality CD → MP3 → Vinyl. My perception is different as I have sound-visual Synesthesia, so all of these formats ‘look’ different to me. All of them colour the sound. Vinyl introduces harmonic distortion and doesn’t suffer from aliasing or loss of intermodulation products like CDs, and noticably improves the top end. Personally I like the sound of Mp3s, although I have to use high bitrates as noise like guitar distortion or hi-hats encodes poorly and looks really quite wrong to me. I can see the gaps in the MP3 audio when one sound has masked another, much like FM radio.

Most people probably won’t notice the difference though. A young undergraduate in his sandwich year whom I met at work recently buys vinyl records, I think he bought them for the artwork and doesn’t have a actually have a record player at all…