Jump to content

Symmetry of TNG

Level1
  • Posts

    146
  • Joined

  • Last visited

Posts posted by Symmetry of TNG

  1. Thanx Everyone!... your greets mean more than you know!

     

    Group Hug, come on =) in a straight, hentrosexual way =)

     

    :hug::hug::hug:

     

     

    GT just got an idea, publish it later as a book ... =) ...name it "asm romance" and put it under surrealistic art and it will sell in millions :P

     

    Thanx 2u all!

     

    /Symmetry

    Yet another revoultion around the sun older...

    Man I'm geting dizzy!

  2. Asshoot... though I commented on this =)

     

    This is just great work man!..

    I never got the GPU debugger option to work with my alpine.. so always debugged code in my head.

    Now i can use it for something else =)

     

    I hope work on this will be continued it is a verry useful tool.

     

    THANK Y-O-U! Orion & coworkers for beeing the F-I-R-S-T one to release such a Great! tool for the jaguar scene! ;)

     

    /Sym

  3. >and the connections are made using inexpensive and easy to find audio cables (no hubs/switches are needed).

     

    on the other hand Hubs/switches & TP cable are almost a standard today and cabling are most probably already done in a house..

    And having a Ethernet connection & perhapps some day an online gaming would open new doors for jag fans.. (that cant go to rgc/jagfiest for a quick game with a friend...could easily game online with them...)

     

    There are actually werry neat, already made things, for "cabled ethernet network" ...

    ...I need to PM SPCD..

     

    Anyway nice to see progress! =) (though i didnt understand that Spy thing... (is it for eavesdropping on communication? isnt that just a debugtool? or is it actually usable for something usable?)

     

    cheers

    /Sym

  4. Nice!... =)

     

    So.. you are saying that its slower to copy data FROM gpu mem 2 dram (5632) than it is to copy TO gou mem(3584)?

    Any ideas why? I dont realy understand that from a logical point of view.. ? ..does the DRAM timings differ in read vs write? ...sram should have same read & write time right? ..how about dram?

    or is it an OP stall, page miss on writes to dram or what? why is it slower to copy TO dram than it is to copy from it?

     

    What are the GPU dooing in your example code? ...Im just curious because I have always wondered what happens when you start a GPU->dram blitt and continue running GPU code afterwards... I mean logically someone would have to halt to give the blitter acceess to gpumem right?... ie a gpu2-dram blitt would halt gpu exec and parallelism would be lost, or?

    Can this be verified with your LA? ....ie faster blitts from gpumem if gpu is turned off. (or perhapps its just a stall issue, once the blitt start it finishes in same time?!)

     

     

    Its nice to see that the +$8000 is realy faster, because in my experiments (iirc the tunnel code) It had no effect at all.. not something that could be seen in fps count anyway.. (or "raster bar timings", they were same size).

     

     

    Last question:

    When i did my own timing test of the memory (simple color "rasters" on screen) the DRAM->DRAM blitt code was As Fast as anything that had to do with the GPU memory!... I assume src&desr was within the same Page in dram hence the speed..

    But it would be nice to se such a timing... ie blitter phrase blitt of dram->dram, preferabely within the same page. ...well offcourse it would be nice to see what happens if blitt crosses page boundary so there will be a page miss on each phrase.. (then perhapps we will se my point on why cleverly written GPU sdram code would be faster than the "running from main ram" thingy..

    Perhapps this is because dram->dram will be true 64bits, while dram->gpumem would have to be 2*32bits..

    Ie keeping the page boundaries in mind when you design your code will give you the ultimate power of the jaguar!.

     

    sorry for all question, but timing issues are interesting from an optimisation point of view :P

     

    Nice work!

    cheers

    /Sym

  5. Thank you all verry much! :hug:

     

    Sorry for not beeing online so much atm.. I'm finding myself building electronics equipment in my free time... dont know who spured that interest? spcd? zerosquare? ...guess you are both "guilty" ;)

    ..well it is for a good cause so I guess i should thank you instead :P

     

    Thanx everyone! I honestly appresiate it!

     

    regards

    /Sym

  6. Hmm a 3D related question out in the open.?.. ok fine by me :P

     

    One can debate forever if it is worth the HOLE lot of extra trouble it is to get pixel perfect polys and compare that to the extra polycount or framerate you get if you just accept some dropouts..

    But i guess for a racing game it doesnt matter that much, but for a "Tomb raider" it might.

    (cf Ps1).

     

    So.. Bresenham.. when i sat down with it and did some comparisons if a line goes from upper right to down left in a polygon, the polygon that gets below it, its rout will fill over large portions of the already drawn upper poly!... ?... for flat polys this might not matter much, you will never get dropouts for sure, but for example if you use textures, you will loose allot of textures from upper poly when the lower one paints them over.

     

    What you need is a "Fill convention" ...you need to descide to have, for example, a "top-left" fill convention, where you "round up" your scanline startingposition on the top & left side of your poly, and round down on bottom& right.

     

    check out Chris Heckers article on Perspective Texturemapping at

    http://www.d6.com/users/checker/misctech.htm

     

     

    Havent realy checked my old rout with a lupe, but i think it looks rather ok (?) ..16.16 math it fits the blitter nicely (with its fractional pixelstart positions). it will automatically deside if the blitter should draw the pixel or not and handle the fraction->screen-integer-pixel-position truncation for you.

     

    Perhapps there is an easier way to do this altogether, if someone knows please let me know =)

     

    cheers

    /Sym

  7. hmm.. anyone feel up for the task of making an "alpine version"?

     

    ...I would like to have such a working environment. since I dont want to mix dev env to much.

    Last time i tried this i ended up partially re-wrighting some version of this code from the HQ cd.. doing this n that and the finally compiling version chewed on my test CD for hours.. then program crashed, Dooh! =)

     

    I admit it, Im ashamed of not having encryption working yet... =) ...but I always felt that "I always have Domins BJL version" =) ...

    Ahh well... now i have a cd version aswell :) odds are getting better n better every day :P

     

    cheers

  8. and the fish shoots what exactly? :P

    sorry naughty naught me =P ...

     

    sincerely... multimedia fusion.. is there a runnable version somewhere? animation says more than a 1000nd screenshots =)

     

    I beleive any game would be nice on the jaguar ..so... start learning to code ;P

    cheers

  9. actually I beleive this to be nice work man =)

     

    Though last time I looked at encryption i ended up trying to get the encryption src (version unknown?) to compile, since I never saw the "true" atari dev kit program available somewhere, ie one that use the alpine setup. Having an alpine thats the code i'd prefer to use (not to mix up to meny dev env's). and afaik you always released bjl versions? ....do you have a working alpine version? (then i'd be verry interested =)

     

    but as before as a last resort I always have your bjl prgs as a way out =) ...and now i have, yet another, (better?!) tool to do it =) thanx

     

    /Sym

  10. sorry itmust be something with my english.. everyone seem to missunderstand me =)

     

    in the pdf file I can read:

    1) under "team tap compatibility" at the end i can read "to prevent damage to your jaguar....." ..."not more than 3".

     

    2) under multi tap comp. i read again.. "to prevent potential damage to your jaguar..."

     

     

    ?.... I never said that connecting 1 controller would destroy your jag, read my post again.. *handen i pannan* ..Nor did i say that I would not include that in "my controller manual"...

     

    NO! wait..... its better to take things once again from the top.

    Some people seem to think that I wanna "shoot down"this project... and that, I consider to be an Insult to my jaguar self (though they are already forgiven for thinking that of me =) ..On the contrary!.. I just thought to say my thoughts Before final production of the units, since I want EVERYONE to have one!..

     

     

    so I will try to reformulate the "destroy your jaguar" statemet as follows:

     

    I would verry much like to see a solution that is "idiot safe" (pardon the expression =) ..ie I would highly recomend rethinking the HW solution to make it NOT possible to ever blow up your jaguar even if you didnt read the manual that says "dont connect more than 3.. or dont do this or that".

     

    Do you follow? ...ie i'd rather see a solution where you either a) make an external PSU or B ) detect the teemtaps (in hardware!) and make the controller useless. ie make it 1controller/jaguar port. (What i ment was that the ideas i had of this project had such a "limitation".. since I beleive it should never be a risk to destroy my jaguar just because i didnt read the fineprint in the manual.

     

    Yes option a) would increase price no point in saying that... but it would also give full rumble pack support.. something I would like! =)

    I sincerely hope that explained my point better.. (?)

     

     

    some comments:

     

    hmm.. 6ms to read &decode 8 controllers...14ms to do other things.. ie 30% time just to read the controller (if your estimations are correct). no wonder the only game that supports it isnt that great :P ...

     

    >does not mean that the game objective or graphics also have to be so advanced to make reading 8 advanced controllers impossible.

     

    No but we could HOPE, could we not? ;)

    I dont want an analoge paddle for PONG.... nor an analoge white man that cant jump =) ... I want to analogusly move the shotgun in Doom or move my head in any other FPS or 3D game... And hence I'd sattle for 2 analogus player/jaguar ...(OR if the 1) option is choosen above.. then this is no diskussion since 8players could be connected without burning some PCB traces on your jaguar... though gameplay would be limited to 8ply pong.. hmm.. =)

     

     

    >could lead to people inadvertantly overloading their Jaguar then you have raised a valid point

    YES!.. =) ..thank you... thats exactly my point.

     

    I can atleast think of one jaguar user that would connect 2teemtaps & connect all ports to such a controller... Without reading the manual.. (since he never read a manual in his life =) ...point beeing that no hardware should offer the possibility to destroy your jaguar (if you by accident didnt read the manual).

    Then I would suggest doing it in another way.. in the HW i mean.

     

     

    15 units:

    Well... I understood what you point out. but my point is that I want to use my old PS controller without sending it 2u =) (if you pardon my honesty). Ie an adaptor would be the easiest way for me, cut&solder would still be childsplay to most people.

    And i cant beleive Im the only one that feels like that? ...or that Im the only one that have an old yes, but working PS controller lying around that could be used... and i mean the adaptor should be so well done that if control fails to work, then i'd be sure it was a broken PS controller and you could just get a "new" one at ebay (or whereever) and there would be no need to re-alter/resend the controller to you for remodification.

     

    I mean from a consumer point of view I would have it no other way.. since that would be a great analogue adaptor product for the jaguar. And from a "hobby" producer point of view I would sugest you to rethink the idea of "warranty"... I mean even if you would offer "limited warranty" on new controllers, even shorter "warranty" on old PS controllers (to fill up possible demand) It might be hell for you when peoples controllers just happen to break JUST after they bought it from you.. "and I want a refound! and offcourse I will not pay shipping!" ...

    I mean.. Im just trying to be helpfull here.. I dont even want my worst enemy (NOT saying you are one) to feel the wrath of unsattisfied customers demanding warranty on their broken unit that they Just bought!.. and "didn't do nothing wrong" with.. *shruggs* =)

    ...I mean please dont go there you'll just be miserable.

     

    It would be much more easy to controll the "warranty" issue with an adaptor approach... since All adaptors could be professionally made and tested in advance. and you could leave the 'finding a 2nd hand PS controller' trouble to the end user and he could never blame you if his controller didnt work perfectly for 3years after purschase.

     

     

    >adaptor firmware recognize & drive appropriately

    Yes it would.. and that is why this is not such a simple task as one first would assume.

    And also why It will take time, effort, hard work to make one.

     

     

    some last comments:

     

    1) ?

     

    2) jagscene do have problem with warn out licence issues I know thats a fakt =)

     

    3) no need for hard words.. I know it is not an easy task, I know it would incorporate some rev. engineering on the protocal, and 13eauros was as much an estimate as your 1ms was ;)... but Im no fool when it comes to electronics and Im perhapps not as stupid as some people might have you beleive :P and its not in the hardware as such but in the program thats written for the adaptor CPU.

     

    4) I did resurch to and desiced its best to make my own investigation not to base my project on other peoples guesses.

     

    5) Nop.. sockets stay... =) btw Im not "blaiming" you for my desision to buy them... though Im assuming this is what you implied (?). You have to realise that I never meen bad with my posts (usually atleast, some jagscene issues drives me nutts but nothing here at JW) and any harsh interpretation is on the reciving side and never ment by the poster =)

     

    Well I hope that you keep us (&me=) informed of the progress anyway... since I would Love to see an analogue controller for the jaguar.

    And No Hard feelings! here at Jagware we are all friends! =)

     

    cheers

    /Symmetry

  11. ...well.. I started a similar thing some year ago..

    but my project had much less "will destroy your jaguar" in the specs..

     

    I like the idea of a analouge gamepad for the jaguar.. and any future projects I will release will defenitely support one, but I beleive he's going about it in a bit of the wrong way.

     

     

    1st Im sure atari never intened more than one, or max 2 intelligent devices to be connected to the jag at the same time (currentconsumption,time to read them all, usually an advanced controller means advanced game -> have the jag got time for more than 2 simultaneous players?,....)

    Hence Id make it detect a team tap & shut down if its present.. (Not to blowUp your jaguar! this should Never!! be an option!). and make max2 controllers / jag, one to each port.

     

    2nd If i got it right, hes going to make a custom made, controller..(alter internals?) and as i hear it will be dificult to find good enough controllers to sell as new products.. ?.. this I beleive will limit the product to "rare" status even before its official release. and it will be useless to jaguar gamers, since no games will be made for a market of only 15 (or so) controllers.

     

     

    Why not instead make it more a "true" adaptor: 15pinHDsub,pcb,CPU,a PS socket + a bit more electronics, at a production cost of... well.. 120SEK Im not exagurating.. (~13euro,8pounds).

    Thas what I planed for my project.. (now I will atleast have ALLOT of PS sockets over for other projects since i bought them already :)

     

    Another more "drastic" idea would be to make the above but without the PS socket.. make it a "cut the coord-solder here" kind of thing. Would destroy your PS controller... but you could take ANY PS controller doing this.. and there is no shortage of ps controllers in general.. and Every jaguar gamer could in theory get one... makeing the market bigger, making people wanna support it in their games...

     

    Both these options could offcourse support other devices aswell.. not just the controller, but also a PS Analogue steeringweel, a lightgun, what more gadgets were released? ....well I also bought a PS memorycard to experiment with, since the PS sockets has a 2nd port underneeth for a memorycard (but thats nolonger a plans due to unlimitless space in the jagCF), ....dance mat?...

     

    Well point beeing that if its not "custom made" there will never be a shortage of hardware to connect!... this I beleive should be strived for! since we do not need "yet another ultra rare object that will never be usable" for the jaguar.... we need "Gameplay-for-the-masses" :)

     

    my thoughts on this topic

     

     

    cheers

    /Symmetry

    supporting: Gameplay for the Masses :)

  12. hmm.. and another thought. dont know how collision detection is done in PA, but i can imagine that if such a bullet engine is done, it could fastly go complex & memoryconsuming if pixel precise collision is wanted.. with for example shots that are not just dots, but bows,fire-bolts or whatever shape based on a bitmap...

    But i guess that is a later problem.... though it might be a good idea to plan ahead....

  13. as i understand it a parser would run through the ascii text and look for (human readable text, like) "action" .."fire" commands and do all that... to me thats a waste of time.. personally id rather see a clever datastructure (in already done dc.b format) for each effect, that calls specific routines that handles the shots... ? (ie convertion to micro-friendly code was already done)..

     

     

    but on the other hand Im not doing a 2D shot-em-up game ..so i should perhapps "shut up" instead ;)

     

    But it was so cool that i just couldn't do that :P

     

    /Symmetry

  14. hmm.. when i first read about it i felt like "is it realy nessesary why not just do this and that in asm"..

     

    then ran the demo.. =) ...this is Supperb for a Shot-em-up!!!!! ...and offcourse a "must" if one wish to make zillion of shots... (and one should strive for that :)

     

    havent looked at the code yet.. so dont know how much work it would be to port it to asm.. though i feel that that would be a must..

     

    Similar thing might be used to make "VLM" style effects... well.. ?.. gaming would have prio 1 offcourse :P ..just saying...

     

    porting:

    ...is it possible in an easy way? ...some text-parsing thing is not my cup of tee, but rather a dc.b + jsr's that do stuff with a list of spritepositions.

     

    wrigting our own:

    well the idea is good.. so one might have to do this to get a "jaguar version".. for mximum speed..

    compatibility woudl perhapps be nice (dont know what one could get for free from that engine.so)

     

     

    Nice Link anyhow!... verry interesting!

     

    /Sym

  15. Jagware will do the Jaguar version of Another World, Eric Chahi (the author) we have meet this week end has gived his permission for that.

     

    The smileys to my left aren't happy enough to express my feelings =)

     

    Another World are one of the few games that i actually played all to the end.... (years ago on my atari) and just loved every minute of it! the moode, the animations everything...

     

    would/will Looove to c it on the jag! :)

     

    thanx to author and all people that made this possible!

     

    /Symmetry

    --Jaguar Future have never looked brighter--

  16. :yes: :yes: :yes: JAGWARE!

     

    You must start planing these events so people working (far far away) have a chance to visit :P

     

    I dont know where else to post this (casual comments) hope it is ok if i do it here:

     

    -SeaPlane looks promising!... I wanna play "1942" on my jaguar ;) ...keep up nice work!

    -ProjectApocalypse ...need i even say im far beyond impressed?! =) ...(Björns gfx ser bra ut ;)

    -CF... more.. More! ...MORE!!!!.. runtime videos, video of debug session, .... a Windows Wallpaper, WHATEVER! *hehe* :P ...where do i send the sponsor money? ;)

     

    Just wanna say that everything looks great, and in the last year or so jagware scene have produced more "actual results" in the UGD jagscene than i have seen in the past decade, software AND hardware wise!!.. ...something, I for one, am glad for and looks positive on!.

     

    *cheers on you all!*

     

    Wish i had been there.. and someday I will!.

     

    regards

    /Symmetry

  17. Thanx!... would have been a pleasure to meet you all too..

     

    Im defenitely going sometime though... a UK one for sure, a frensh one hopefully!.

    It just require more timing and planing =) ..

     

    Yes I hope there will be a video demonstration of the JagCF ....similar to the one done for the JagVR ;) ...(but better interview offcourse :) ..would be interesting indeed..

     

    The jaguar will live forever, so cu another time!

    cheers

×
×
  • Create New...