Author Topic: Biggest lost opportunities for the PCE  (Read 5015 times)

Arkhan

  • Hero Member
  • *****
  • Posts: 14142
  • Fuck Elmer.
    • Incessant Negativity Software
Re: Biggest lost opportunities for the PCE
« Reply #75 on: May 17, 2011, 03:23:57 PM »
I think the SGX didn't offer enough to encourage development when you have massive CDs to throw games on.  I mean, Cosmic Fantasy 3, cmon.
[Fri 19:34]<nectarsis> been wanting to try that one for awhile now Ope
[Fri 19:33]<Opethian> l;ol huge dong

I'm a max level Forum Warrior.  I'm immortal.
If you're not ready to defend your claims, don't post em.

spenoza

  • Hero Member
  • *****
  • Posts: 2751
Re: Biggest lost opportunities for the PCE
« Reply #76 on: May 17, 2011, 05:01:36 PM »
I think the SGX didn't offer enough to encourage development when you have massive CDs to throw games on.  I mean, Cosmic Fantasy 3, cmon.

Ah, but what if you could have both? Seriously! The limited CD RAM would still have been a limitation, but having more VRAM and main RAM available could have been a very good thing.
<a href="http://www.pcedaisakusen.net/2/34/103/show-collection.htm" class="bbc_link" target="_blank">My meager PC Engine Collection so far.</a><br><a href="https://www.pcenginefx.com/forums/" class="bbc_link" target="_blank">PC Engine Software Bible</a><br><a href="http://www.racketboy.com/forum/" c

TheOldMan

  • Hero Member
  • *****
  • Posts: 958
Re: Biggest lost opportunities for the PCE
« Reply #77 on: May 17, 2011, 05:35:52 PM »
Quote
having more VRAM and main RAM availablecould would have been a very good thing.

Yeah, with more VRAM, we could load a bigger BAT and more graphics to boot. Larger screen, better animation on larger characters. <sigh> More main memory would allow tracking more objects, and probably better AI for objects.
What could have been......

Bonknuts

  • Hero Member
  • *****
  • Posts: 3292
Re: Biggest lost opportunities for the PCE
« Reply #78 on: May 18, 2011, 03:09:59 AM »
Missed opportunities....

 There are a ton, hardware wise. We know from articles and interviews that the PC-Engine itself was always destined for more than what it was released as. The PCE by itself was nick named the 'core' system. The CD addon was in development and show cased even before the system was officially released. But onto specifics.

 - Lack of adequate core ram. More than just speculation, it's believed that the PCE original was developed with 32k of ram in mind, but only shipped with 8k. The upper 24k is just the 8k mirrored, not open bus. Fittingly, the SGX actually populates this mirrored ram with the additional 24k to make the original 32k ram. 8k of ram is limiting. It limits the type of compression schemes you can use. It limits caching decompressed data for later parts of levels (instead of doing it real time).

 - The cut back of vram to 64k. The VDC has full access to 128k of vram. The arcade Bloody Wolf which is basically just a PCE without a VCE, has that full 128k. It has all the player and enemy frames loaded in. Having larger vram space also means you can use better compression schemes and cache the frames into vram (like you would normal ram). I would have put more emphasis on this than regular system ram. Especially when you consider the CD addon original only gave 64k of work ram (simulated cart rom).

 - The lack of any second BG layer. Scrollable or fixed overlay/underlay. The PCE is setup to have a digital pixel bus to another device. That type of setup is used in quite a bit of arcades system, to mix different layers of different video chips. In the context of the PCE, the VCE is almost a waste. Bloody Wolf gets by just fine without it. It makes me wonder if something else was planned but cut. Maybe the SGX was closer to the original PCE design, but was cut down due to costs.

 - Audio. It's not surprise to me that the stock audio isn't miles ahead of the NES or even home computers at the time. The CD addon was more than likely to be their key marketing advantage. The difference of chips tunes like the PCE has, to that of CDDA audio - only strengthens the CD addon's value. There are upgrades that can be done via cart (mono input being one of them), but also a higher res timer. You can use the scanline interrupt (all 262 lines) to make 16khz audio, but it's rather expensive. An external timer that you resync on vblank INT would be perfect. It could even digitally mix channels for you and output them on two 5bit channels (10bit output), etc. But of course, hucards were replaced with dominant CD format fairly earlier on. The real mist opportunity is on CD games. You have 6 channels to use at once to make some really unique and great sound FX, yet more settled back on generic/typical PSG-ish tiny type SFX.

 - CD addon memory. The original CD addon memory was DRAM. DRAM is/was much less expensive than SRAM. Yet they only included a measly 64k. With only 64k of vram, 64k of simulated cart space is pathetic. They should have gone at least with 128k. At least that. It ended up costing them more in the long run because the there is no /RDY for DRAM interface on the cart port (on the expansion bus), so the 192k upgrade ram on the SCD card is SRAM. Much more expensive than DRAM. They should have started out with 192k or 256k of DRAM to begin with. I mean, considering the cost of everything else in the CD addon base ( the MCU, the cd drive logic, the ADPCM IC + ADPCM ram) - it would have been relatively low.

 - VDC. The VDC in the PCE (basically THEE video chip) spends quite a bit of time doing nothing during active display. IIRC, it spends have the scanline doing nothing. It gives all those slots to the CPU. For an engineering standpoint, that's fairly wasteful. There's totally enough time to actually fetch and build another BG layer there. But more important than that, is how amazing fast the VDC fetches all the sprite pixels. It does it in the short amount of time of hblank. No other system does that. Other other systems from that era use the whole scanline to parse and fetch the sprite pixel data for the next line. Not the VDC, does it all in hblank. But here's the kicker... there's only 64words of memory for sprite line buffer. That's 256 4bit sprite pixels. The VDC is fast enough to fetch much more than that, and it tries - but there's no more space to store them so they get dropped. Considering they chose not to have a second BG layer or static under/overlay window, they should have AT LEAST increased this to double the size. They had to have known that developers were going to fake BG overlay parts with sprites. Making sprites all the more important. Plus, that means the sprites to scanline ratio with also scale with the increase resolutions (mid res, high res). But because of this limited buffer, it doesn't. That's more said than any missing BG layer. Such power wasted and crippled by a small internal buffer. To me, that's the single best mist opportunity of the system design.

 - Yes, the SGX could have been made an addon. But not as it is. The extra 24k of ram sits where it's normally mirrored on the PCE. If you try to make ram there either via cart or expansion bus, you'll get a bus conflict. Just won't work. But everything else, yes. But there's a simple solution for SGX software to see if it's running on the main console or as an addon. It tests the mirror ram, if present then it used alternate bank mapped ram. Simple as that. I came up with this idea because the CD system card does just this. There's a reserved bank number in the last few bytes before the vector address of the system card rom. CD software take this value and use it as the starting bank for CD ram. Oddly enough, Gate of Thunder actually tries to change this byte in rom. From what Charles MacDonald told me of the CD dev card, system rom and ram are in different places - so this starting bank number was put there for a reason. GOT have some left over dev card routine to change it, but since it's rom it doesn't change. Anyway, the very same method could have been used to play SGX games on the main SGX system or the addon SGX upgrade.

 - CD addon. The CD addon was in development when the PCE was too. Though the PCE was release first (and a year apart IIRC). The initial faults of the PCE should have been clear in the first few months. The CD addon could have addressed some of these. Adding a second VDC to the CD addon is VERY doable. It doesn't even have to be as extravagant as the SGX. The original VCE output would be ignored and a new VCE inside the CD addon could have upgraded the RGB resolution to 12bit too. Or some other scheme ( a 10th and 11th bit for half intensity and double intensity in the RGB entry), etc turned on by a special reg otherwise left disabled for compatibility reasons.

 Adding the video upgrades, even if less than what I suggested (maybe just a static overlay/underlay 8bit image and a new VCE palette expansion) - would have made the CD addon stand out even more. And when the Duo was released, it would no problem matching the level of the SFC and MD hardware wise.

 Sega f*cked up. They didn't have even half what the PCE had for an expansion bus. Yet they tried everything they could to have the MegaCD more advanced. It was... no it IS hackish. If you even dev'd on the SegaCD then you know exactly how hackish it is. The PCE had all the right pins and access lines needed to make a real hardware addon.... and they squandered that opportunity. They squandered the opportunity to make the CD addon and incredible upgrade that it could have been, not just some addon with a few frills here and there.  :(

soop

  • Hero Member
  • *****
  • Posts: 2828
Re: Biggest lost opportunities for the PCE
« Reply #79 on: May 18, 2011, 03:37:02 AM »
Reading that is just incredible, but it kind of makes me sad.  I need you to go back in time :(

Arkhan

  • Hero Member
  • *****
  • Posts: 14142
  • Fuck Elmer.
    • Incessant Negativity Software
Re: Biggest lost opportunities for the PCE
« Reply #80 on: May 18, 2011, 06:42:27 AM »
The PCE's PSG is miles ahead of the other chips of the time.  Save FM and SID, but those are two very different concepts. 

I rather enjoy stereo panning on 6 channels, with 32 byte waveforms and sampleable nonsense.

Beats the piss out of the NES, or the MSX, Spectrum, Coleco, Intellivision, all the Atari's, the Master System.... the Atari ST....

it just kicks ass.


I'm glad they never got rid of it all the way through the PC-FX basically.  It's a unique and very fitting sound chip for the era. 


They should have just released a GrafxBooster

thats what it should have been called.
[Fri 19:34]<nectarsis> been wanting to try that one for awhile now Ope
[Fri 19:33]<Opethian> l;ol huge dong

I'm a max level Forum Warrior.  I'm immortal.
If you're not ready to defend your claims, don't post em.

Bonknuts

  • Hero Member
  • *****
  • Posts: 3292
Re: Biggest lost opportunities for the PCE
« Reply #81 on: May 18, 2011, 11:44:16 AM »

Quote

the MSX, Spectrum, Coleco, Intellivision, all the Atari's, the Master System.... the Atari ST....

 Those sound chips are basic toy chips IMO. My comment about the chip is relative to the NEC PC line (being the most popular in Japan and being of the same developer... NEC). All FM chips. Even the lowest grade ones too. The PCE chip lacks any form of timbre control on a single channel. You have to pair channels just to try and emulate the effect. That's one of the reasons why you still see comments to this day, that PCE chip music sounds like NES-ish stuff. Yeah, it's much more capable than the NES overall. But it wouldn't have hurt them to make it an even 8 channels. Making channel pairing more viable option as single pseudo channels. The lack of timbre control like real music synths or even low grade FM chips, sets the PCE chip into lower category. Or hell, something even more simpler than that. Something so simple, it makes me want to punch the PCE audio developer in the face. A simple read back of the waveform pointer position of the f*cking selected channel. "I mean, come on!" (arrested development reference). That alone would have made the PCE audio chip into a real wavetable synth (no, not sample based synth like the SNES/MOD/ETC). That one simple little feature. Bloody Wolf developer said "f*ck it, were gonna do it anyway", and tried to do it with the trumpet instrument - but without the knowledge of the waveform pointer it comes off as course and click-y (for obvious reasons).

 As for the PCFX, that's the most moronic thing they could have ever done. Why the hell would they do that? A 1994 system (might as well be 1995) with an audio chip from 1987??? WTF were they thinking? It should have at least had an 8 channel sample based DSP and an 8 channel Yamaha 2151 (standard stuff for that era). It might be cool or unique now, but BITD it was f*cking laughable. Of course, there are a lot of things about the PCFX design that just boggle the mind. IF it was released AS IS in late 1992... then maybe it would have been appropriate (video and audio specs). But 1995 is a joke. To me, the PCFX is a SuperGrafx 2.0.


Arkhan

  • Hero Member
  • *****
  • Posts: 14142
  • Fuck Elmer.
    • Incessant Negativity Software
Re: Biggest lost opportunities for the PCE
« Reply #82 on: May 18, 2011, 12:09:04 PM »
Oh.  Well yeah of course its dwarfed by the PC-88/98 line.  When you have a computer that has plenty of room to expand, its real easy to toss in a pretty powerful synthcard.  FWIW, the original FM card in those things blew pretty bad though.  Its like 3 channels of FM only. and 3 SSG.  So, its pretty bleh.  Since it is a computer though... lots of stuff was able to come out and upgrade that into the sounds the thing is really known for.   Doing this with the PCE would have either made it cost more, be larger, who knows.  I'm glad they picked what they picked.  We've got enough f*ckin FM crap out there.  The PCE is pretty damn unique in sound land, mimicked only by the SCC, and the NES w/ add ons.    And neither of those do it as good.

Some of the old PC-88 games have some pretty shit music. 

I think the PCFX's sound choice is a nice acquired taste.  You have to remember, there is CD audio.  With CD Audio, you can dwarf the entire PC-88/98 line.  You can record just about anything.  45 FM synths playing in tandem recorded in a studio if you want.  Suck on that, JAST!

They probably left the good ol' chirpy chip in there as a throwback.  "Yeah we got badass CD audio, but we still have your nostalgic sounds too".

I like the music it pumps out for Yuna.  It sounds really cool.

Thats the beauty of CD though... I used a SID for Insanity's music... ;) 
[Fri 19:34]<nectarsis> been wanting to try that one for awhile now Ope
[Fri 19:33]<Opethian> l;ol huge dong

I'm a max level Forum Warrior.  I'm immortal.
If you're not ready to defend your claims, don't post em.

ccovell

  • Hero Member
  • *****
  • Posts: 2245
Re: Biggest lost opportunities for the PCE
« Reply #83 on: May 18, 2011, 01:32:29 PM »
The PCE is pretty damn unique in sound land, mimicked only by the SCC, and the NES w/ add ons.    And neither of those do it as good.

The SCC might be mono and only 5 channels, but it's much deeper and richer than the PCE's sound.  8-bit samples vs 4-bit samples does make a difference.

I agree that the PCE's PSG is a bit simpler and less ambitious than FM chips, but it still has a good sound to it.  Also, Hudson did the development of the whole PCE core chipset, so NEC wouldn't have had any say (or relevance w/r/t their line of PCs) until after Hudson started shopping around for manufacturers.

fragmare

  • Hero Member
  • *****
  • Posts: 676
Re: Biggest lost opportunities for the PCE
« Reply #84 on: May 18, 2011, 02:24:58 PM »
On this same topic, a small processor chip in the 1-2 MHz range for handling the PSG would have been nice.  As it stands, the Hu6280 CPU handles the PSG sound, and it eats up 5-10% of processing time that could be freed up for game code.  Not a huge deal, but still would have been nice.

Arkhan

  • Hero Member
  • *****
  • Posts: 14142
  • Fuck Elmer.
    • Incessant Negativity Software
Re: Biggest lost opportunities for the PCE
« Reply #85 on: May 18, 2011, 02:30:55 PM »
The PCE is pretty damn unique in sound land, mimicked only by the SCC, and the NES w/ add ons.    And neither of those do it as good.

The SCC might be mono and only 5 channels, but it's much deeper and richer than the PCE's sound.  8-bit samples vs 4-bit samples does make a difference.

PCE is 5-bits?

The SCC might be deeper/richer (usually the case with Konami games, not always the case with independent use), but not having stereo panning really does suck.  Especially when you hear alot of the stereo stuff on the PCE, like Dungeon Explorer. 

Plus the drums are generally done with the regular MSX PSG, which is better or worse, depending how you look at it, and whos doing the drums.

All the MSX people I've talked to really envy the per channel panning. 

[Fri 19:34]<nectarsis> been wanting to try that one for awhile now Ope
[Fri 19:33]<Opethian> l;ol huge dong

I'm a max level Forum Warrior.  I'm immortal.
If you're not ready to defend your claims, don't post em.

spenoza

  • Hero Member
  • *****
  • Posts: 2751
Re: Biggest lost opportunities for the PCE
« Reply #86 on: May 18, 2011, 04:52:05 PM »
Sega f*cked up. They didn't have even half what the PCE had for an expansion bus. Yet they tried everything they could to have the MegaCD more advanced. It was... no it IS hackish. If you even dev'd on the SegaCD then you know exactly how hackish it is. The PCE had all the right pins and access lines needed to make a real hardware addon.... and they squandered that opportunity. They squandered the opportunity to make the CD addon and incredible upgrade that it could have been, not just some addon with a few frills here and there.  :(

This is the story of every piece of Sega hardware after the Genesis up until the Dreamcast. The Dreamcast was a hardware jewel, but the Saturn, 2D beast that it was, was a real hack-job.

I wonder if all these hardware possibilities were part of Hudson design works, or if they were due to NEC's involvement, or if NEC's involvement prevented them from pursuing those avenues. You wouldn't want the PCE, as a CD unit, begin to usurp PC-88 and, later, PC-98 sales. The CD-ROM was computer-bound...
<a href="http://www.pcedaisakusen.net/2/34/103/show-collection.htm" class="bbc_link" target="_blank">My meager PC Engine Collection so far.</a><br><a href="https://www.pcenginefx.com/forums/" class="bbc_link" target="_blank">PC Engine Software Bible</a><br><a href="http://www.racketboy.com/forum/" c

Arkhan

  • Hero Member
  • *****
  • Posts: 14142
  • Fuck Elmer.
    • Incessant Negativity Software
Re: Biggest lost opportunities for the PCE
« Reply #87 on: May 19, 2011, 09:50:10 AM »
Wasnt the PCE eventually intended to become a fully functional computer setup?

Maybe NEC helped put the brakes on that.

I think back then, they also planned for things, and didn't plan for others...... because this was all a new frontier back then. 

Hindsight is always 20/20.  A lot could've been done differently.  They could have done video addons, sound addons, all kinds of crap that we see plain as day right now...

Back then though, everyones favorite retro sound preferences weren't retro yet.  They were still becoming preferences.  The preferred format wasn't decided on (until the Playstation came and went "Yep, welcome to CDs. If you don't follow suit, you suck.  You hear that Nintendo? You suck.").

Lots of things were still experimental.

screw the missed opportunities anyway.  PCE was king back then, and it still is.
[Fri 19:34]<nectarsis> been wanting to try that one for awhile now Ope
[Fri 19:33]<Opethian> l;ol huge dong

I'm a max level Forum Warrior.  I'm immortal.
If you're not ready to defend your claims, don't post em.

Bonknuts

  • Hero Member
  • *****
  • Posts: 3292
Re: Biggest lost opportunities for the PCE
« Reply #88 on: May 20, 2011, 04:16:53 AM »

The SCC might be mono and only 5 channels, but it's much deeper and richer than the PCE's sound.  8-bit samples vs 4-bit samples does make a difference.

 If the waveforms are short, 8bit depth does nothing for it. And then take that into the analog output world (because sample output isn't stair cased like when you look at it digitally), it means even less. The sample lengths for these DDS samples are way too short for the different of 8bit to 5bit to mean much of anything. If you don't believe me, try it yourself in an XM tracker. Make a 32sample 8bit wave and make an identical one in 5bit steps (padded to 8bit though). As it is, you can't even use all 5bit value range resolution for a complete sine wave in just 32 samples. If anything makes the MSX with SCC more rich, it would be the inclusion of the original on board MSX sound chip in conjunction with SCC chip.


Quote
I agree that the PCE's PSG is a bit simpler and less ambitious than FM chips, but it still has a good sound to it.  Also, Hudson did the development of the whole PCE core chipset, so NEC wouldn't have had any say (or relevance w/r/t their line of PCs) until after Hudson started shopping around for manufacturers.

 The over all design is good, IMO. But since it lacks any timbre control per intra-note (you can do it per note level itself though), it really holds it back. Early 70's synths that weren't FM used lots of channels to create instruments sound models, but to also control the timbre of a instrument sound. For the PCE's audio design, it's clearly lacking in the number of channels. By the time the PCE was out, audio chips in game systems were getting more serious (and arcades too). Famicom really stepped up the bar. PCE audio, if nothing changed in design, should have been eight to ten channels, really. Even if some higher channels had to share the same sample buffers as lower channels to cut cost. But then again, if NEC really wanted to 'fix' the issue, the mono audio in line is right there on the cart port. Famicom had a ton of audio upgrades via cart. So it's definitely not unreasonable that PCE wouldn't have gotten them either. But I think it goes back to the CD rom. The difference from chip audio to CDDA is drastic back then (an obvious selling point). If the CD addon had completely flopped, then I'm pretty sure we would have seen some audio upgrades on hucards.



Quote
I wonder if all these hardware possibilities were part of Hudson design works, or if they were due to NEC's involvement, or if NEC's involvement prevented them from pursuing those avenues. You wouldn't want the PCE, as a CD unit, begin to usurp PC-88 and, later, PC-98 sales. The CD-ROM was computer-bound...

 Think of it this way. The PCE wasn't an open development system like the NEC PCs. It required a license (which is where NEC made money on each title/copy sold) and a special dev unit for it. If it was anything like the US, NEC wasn't receiving any licensing fees for game softs created on the PC series. And while the NEC PCs were very popular, I doubt even a 1/4 of the sales were strictly from gamers wanting to play PC games. I think that was more of an added side bonus for kids that got a PC. NEC had more to gain and little if anything to loose by making the PCE game system more powerful than the PC line for gaming. But then again, I didn't grown up in Japan and all of that is based on my personal experience with PC's here in the US BITD.

lukester

  • Guest
Re: Biggest lost opportunities for the PCE
« Reply #89 on: January 16, 2012, 08:53:16 AM »
I know this is kind of late, but anyway, here is my list of missed oppurtunities ](*,)

1. More Capcom and Konami games, a Megaman game and a Contra would have been cool.
2. Namco CDs. ALL of Namco's games were on Hucard. Some ports were fantastic(Galaga 88, Genpei Toumaden), but others were awful(Youkai Douchuuki, WONDER Momo). Splatterhouse had none of the gory backround details from the arcade, and could have been much better on CD format.
3. Supergrafx-no brainer. Come on, 2 out of the 5 games were Capcom games! We could have had some major Capcom support!
4. Direct Toaplan support, they kept developing for the Megadrive :-k
5. Using Nintendo's policy to steal major companies(until it was illegal) before Nintendo did
6. A port of R-Type II