what if the cpc+ had 32 hardware sprites and they had 256 colours..

[TotO]

I don't see a need for 256 colour sprites, but I think having different 16-colour palettes for different sprites would have been very nice.

Absolutely, like most existing systems. 4bit colors + 4bit palettes as exemple.

[ivarf]

And the most fun thing is when Sugar saved the british pride buying sinclair, they threw away the Loki project (Loki -> Konix Multisystem -> Atari Panther -> Atari Jaguar), a lot of that technology could have been used in the CPC family.

All of these machines were bigger failures than the CPC+!!!

[SyX]

It's not about the failure, it's about a nice technology that amstrad got free and instead of using or studying it, they sent to the trash the first day without take a look.

[McKlain]

I can imagine the situation:


-Mr. Sugar: Can we sell this?
-Staff member: Well, not yet.
-Mr. Sugar: Fuck it.

[ivarf]

It's not about the failure, it's about a nice technology that amstrad got free and instead of using or studying it, they sent to the trash the first day without take a look.

I have been looking around and it seems that the techology wasn't there when Amstrad took over. It was developed later by Flare. The Jaguar was released in desember 1993, 7 years and 9 months after amstrad bought Sinclair. Most people at Sinclair and Amstrad at the time considered the Loki to nothing more than a wishlist that would have needed years of development. Some of this seems to have been used for the Sam Coupes graphics, but I can't find much there that Amstrad couldn't have done on their own. What the CPC+ doesn't have has probably more to do with cost and compability issues than what was possible for Amstrad.

[MacDeath]

Ok...


=ASIC was screwed by the fact it included the ACID system management... This used some space inside it that could have been certainly used for more DMAs or Palettes or Sprites or clever RAM management or extra colours attributes or Whatever.

And the most fun thing is when Sugar saved the british pride buying sinclair, they threw away the Loki project (Loki -> Konix Multisystem -> Atari Panther -> Atari Jaguar), a lot of that technology could have been used in the CPC family.

was the "Loki" Z80 based ?


ok what Sinclair could actually offer...


=Zx speccy : was exploited by Amstrad as it used a lot of common component with CPC and PCW... (being a Z80...) Saddly the CPC wasn't upgraded into a full 128k range.


=Sinclair QL : a 68008 based computer... Amstrad produced none of those. Also QL was screwed by hugeload of bad technological choices, such as freaking microdrive, the Video was even almost inferior to what a CPC could do. Too many new componnents to buy or too many research to make this QL like it should be...
And "Amstrad QL" would be actually some sort of Atari ST with 3" disk drives and only 128K RAM...


=PC200 (or PC20 ?) : an "IBM PC" compatible machine in CGA put into an Atari ST casing. Was actually released and was utter failure. Could sort of have worked provided it included a custom video standard (like some CPC upgraded into EGA like power) and a little AY for sound too and even some MIDI port.
Hell even the PC1512 and PC1640 were somewhat better.






What I don't understand is that the Amstrad PLUS should have been somewhat compatible with PCW, CPC and Speccy alongside a super "PLUS" mode... to be perfect. With 2 ASICs and different cartridges OS it could really have been manageable.

[Gryzor]

=Sinclair QL : a 68008 based computer... Amstrad produced none of those. Also QL was screwed by hugeload of bad technological choices, such as freaking microdrive, the Video was even almost inferior to what a CPC could do. Too many new componnents to buy or too many research to make this QL like it should be...
And "Amstrad QL" would be actually some sort of Atari ST with 3" disk drives and only 128K RAM...  

It looked damn awesome though. I love mine (even though it had developed a problem with keyboard. I was too heart-broken to look into it )

[emuola]

It looked damn awesome though. I love mine (even though it had developed a problem with keyboard. I was too heart-broken to look into it )

You got a new membrane for your QL Gryzor? I did and it works fine

[Gryzor]

Quote from: Gryzor on Today at 11:02:27It looked damn awesome though. I love mine (even though it had developed a problem with keyboard. I was too heart-broken to look into it )
You got a new membrane for your QL Gryzor? I did and it works fine

To tell you the truth, no. As I said, I was really sad when it broke down and, combined with a lack of time I never looked into it. How much do they go for?

[emuola]

To tell you the truth, no. As I said, I was really sad when it broke down and, combined with a lack of time I never looked into it. How much do they go for?

Seems to be £14 here:

New Sinclair QL Keyboard Membrane

[Gryzor]

Cheers, much appreciated. I wonder if it's the membrane indeed or something else, though it surely looks like it (IIRC the left half of the keyboard doesn't work). 

[emuola]

Cheers, much appreciated. I wonder if it's the membrane indeed or something else, though it surely looks like it (IIRC the left half of the keyboard doesn't work).

You're welcome I'm 99% sure your problems with the QL keyboard are caused by the faulty membrane. If/when you decide to open the computer you'll be amazed how crappy the membrane actually has turned
into during the years. It's like papyrus

[MacDeath]

perhaps you can refit a QL mother board inside a CPC6128 unit ?

[Gryzor]

Well, actually I had taken a look into the membrane, and it looked ok. I looked at it because the darn thing stopped working after I first opened it up. I'm not sure, but I seem to recall that opening it up puts a strain on the membrane/ribbon?

@MacDeath: why would you?

[emuola]

Well, actually I had taken a look into the membrane, and it looked ok. I looked at it because the darn thing stopped working after I first opened it up. I'm not sure, but I seem to recall that opening it up puts a strain on the membrane/ribbon?

@MacDeath: why would you?

@ Gryzon: Yes, opening an old QL will mostcertainly break the ribbon(s) that lead from the actual membrane to the mb. I know, because that's what happened to me. The original membranes have become really fragile, but the ones I posted a link are manufactured recently=way more durable. The original design of how the membrane is fitted is really awful, because it puts a real strain on the membrane (the part that's hooked up tp mb gets bent pretty badly due to space restrictions inside the case).

[emuola]

I guess this is kinda off-topic, but anyway...

I've always been fascinated about the original PC-Engine/TG-16 design: It's as early as 1987, very compact and yet very powerful (if you ask me). Of course it does not have a floppy/keyboard/serial etc, but as pure cpu/graphics combo its pretty neat. I know the amount of RAM is pretty lame, but combined with HuCard as game media, it did not stop from creating some really impressive games. More tech savvy people here can give their insights on this machine, but I love it anyway Combined with Everdrive (sd-adapter) its a real 80's game dream machine.

Here's are the specs (from wikipedia):

 
  The American TurboGrafx-16 console with CD unit   

   
• CPU: 8-bit HuC6280A, a modified 65SC02 (a separate branch from the 65C02, of the original MOS 6502) running at 1.79 or 7.16 MHz (switchable by software). Features integrated bankswitching hardware (driving a 21-bit external address bus from a 6502-compatible 16-bit address bus), an integrated general-purpose I/O port, a timer, block transfer instructions, and dedicated move instructions for communicating with the HuC6270A VDC.
• GPU: A dual graphics processor setup. One 16-bit HuC6260 Video Color Encoder (VCE), and one 16-bit HuC6270A Video Display Controller (VDC). The HuC6270A featured Port-based I/O similar to the TMS99xx VDP family.

Display  Resolution

   
• X (Horizontal) Resolution: variable, maximum of 565 (programmable to 282, 377 or 565 pixels, or as 5.37mhz, 7.16mhz, and 10.74mhz pixel dot clock)^[18] Taking into consideration overscan limitations of CRT televisions at the time, the horizontal resolutions were realistically limited to something a bit less than what the system was actually capable of. Consequently, most game developers limited their games to either 256, 336, or 512 pixels in display width for each of the three modes.^[19]
• Y (Vertical) Resolution: variable, maximum of 242 (programmable in increments of 1 scanline). It is possible to achieve an interlaced "mode" with a maximum vertical resolution of 484 scanlines by alternating between the two different vertical resolution modes used by the system. However, it is unknown, at this time, if this interlaced resolution is compliant with (and consequently displayed correctly on) NTSC televisions.
• The majority of TurboGrafx-16 games use 256×239,^[18] though some games, such as Sherlock Holmes Consulting Detective did use 512×224. Chris Covell's 'High-Resolution Slideshow' uses 512×240.

Color

   
• Depth: 9 bit
• Colors available: 512
• Colors onscreen: Maximum of 482 (241 background, 241 sprite)
• Palettes: Maximum of 32 (16 for background tiles, 16 for sprites)
• Colors per palette: 16 per background palette (color entry #0 of each background palette must be the same), and 15 per sprite palette (plus transparent, which is displayed as an actual color in the overscan area of the screen)

Sprites

   
• Simultaneously displayable: 64 on-screen, 16 (256 sprite pixels) per scanline
• Sizes: 16×16, 16×32, 16×64, 32×16, 32×32, 32×64
• Palette: Each sprite can use up to 15 unique colors (one color must be reserved as transparent) via one of the 16 available sprite palettes.
• Layers: The HuC6270A VDC was capable of displaying one sprite layer. Sprites could be placed either in front of or behind background tiles by manipulating a bit which caused indirect pixel color entry #0 of the background tile(s) to act as transparent.

Tiles

   
• Size: 8×8
• Palette: Each background tile can use up to 15 unique colors via one of the 16 available background palettes and 1 shared color (BG color #0) for a total of 16 colors per tile. The first color entry of each background subpalette is ignored. Instead, color #0's RGB value is shown in its place (the common/shared color). When a specific sprite is set to show behind the BG layer via the priority bit, all tiles that use relative color #0 (of 16) will not show BG color #0. But instead will show the sprite pixel (if not opaque).
• Layers: The HuC6270A VDC was capable of displaying one background layer.

Memory

   
• Work RAM: 8 kB
• Video RAM: 64 kB

Audio capacity

   
• 6 Mini-Wavetable stereo audio channels, programmable through the HuC6280A CPU.
• Each channel had a frequency of 3.58Mhz PCM sample clock (while not in D/A mode) with a bit depth of 5 bits. Each channel also was allotted 20 bytes (32×5 bits) of RAM for sample data.
• The waveforms were programmable so the composers were not limited to the standard selection of waveforms (square, sine, sawtooth, triangle, etc.).
• The first two audio channels (1 and 2) were capable of LFO when channel #2 was used to modulate channel #1. This was used to achieve FM-like sound qualities.
• The final two audio channels (5 and 6) were capable of Noise generation.
• Optional software enabled Direct D/A which allows for sampled sound to be streamed into any of the six PCM audio channels. When a channel is in D/A mode the frequency is as fast as the CPU can stream bytes to the port, though in practicality it's limited to 6.99 kHz when using the TIMER interrupt with the smallest loop setting (1023 cpu cycles). Additionally, a programmer could use the scanline interrupt to generate a 15.7khz interrupt system to play samples.
• Each channel has its own DAC and two layer attenuation device (two volume mechanism controls) allowing a combination of two channels in Direct D/A mode to be paired and play back 8-bit, 9-bit, or 10-bit linear PCM samples.
• Each channel has a 4bit left and 4bit right fine pan volume register for stereo volume control. The audio unit also contains a master 4bit/4bit pair fine pan volume control, used to set volume/stereo level for all channels as a whole.
• The addition of the CD-ROM peripheral adds CD-DA sound, and a single ADPCM channel to the existing sound capabilities of the TurboGrafx-16.

[TotO]

Sure, we can all check the wikipedia page.

On a videogame system, you don't need so much RAM as all is running from the ROM, except for computing... It's why most 8/16bit video game system don't exceed the 64K 16bit memory adressing. (8/16K is already great)

Hudson made first this technology to improve it's own games on Famicom/NES system by adding chips inside the games cartridges.  (like Nintendo, Capcom and Konami does with poor add)
But Nintendo disagree because it was too much powerful and they plan to release the SNES later... So they found NEC to produce a stand alone system... Yes, the "Playstation" story was not the first Nintendo's fail...

Spoiler: ShowHide

Then, Hudson fail with the SuperGrafx because they keep the same NEC 8bit CPU instead of using Sharp 68000 CPU. (they does for them the X68000 OS and intend to use it on this new videogame system, but NEC disagree to use this concurent CPU) Hudson stay with NEC because the mistake was to call the 8bit system NEC PC-Engine and not Hudson PC-Engine...
So, no 16bit Sharp SuperGrafX but poor Nec 8bit SuperGrafX...

what if the cpc+ had 32 hardware sprites and they had 256 colours..

They had been the slowest 8-bit computer ever.

[Gryzor]

@ Gryzon: Yes, opening an old QL will mostcertainly break the ribbon(s) that lead from the actual membrane to the mb. I know, because that's what happened to me. The original membranes have become really fragile, but the ones I posted a link are manufactured recently=way more durable. The original design of how the membrane is fitted is really awful, because it puts a real strain on the membrane (the part that's hooked up tp mb gets bent pretty badly due to space restrictions inside the case).

Thanks! Will be locating my QL in all the boxes and getting a new one...

[MacDeath]

I looked at it because the darn thing stopped working after I first opened it up.

Weren't all those Sinclair computer reputed for a massive lack of reliability and bad quality control ?


So it can fail in many parts.


PCengine :
Guys you're talking about an electronic masterpiece...
NEC was far better than Amstrad concerning design.


The CPU is largely "overclocked", the Video and sounds chips are also largely superior to most "8bit" things.
I even like to think PCengine could do a great computer if provided with 512K RAM and a sweet Disk Drive, HDD and CR-ROM and sweet OS.
SegaMD would also do a freaking great computer, with 512k of proper RAM, the CD option and so on.

On a videogame system, you don't need so much RAM as all is running from the ROM, except for computing...

Well, the CD-ROms actually had to add a "huge" load of RAM to buffer from the CD.
So puting so few RAM on those systems was a setback in the long term.


But ok, most of those consoles had actual VRAM and quite very few "normal RAM".
still as Cartridges were expensive and limited technology, having more RAM to buffer, do some manipulations (swap/flip sprites or tiles, change a few colours) or manipulate numbers and values, decompress some datas or do 3D or whatever the computers could do more "easily".


Those console were quite handicaped concerning procedural graphics hence the too often classic "arcade 2D tiles and sprites engines".


SegaMAsterSystem also suffered a few from having too short RAM... and some kind of games weren't that often ported on Consoles unless you add extra chips on the Cartridge.

Also supergraphix was nice but actually too huge on graphics so the CPU could have a few problems...
(SuperNES was also too huge on graphics so it limited some games to 1player modes only).


At one point the GX4000 was flawed because it had no way to be upgraded with Tape/Disk and keyboard IMO...


But the 64K RAM was actually needed because it was also used as "VRAM" and a fullscreen/overscan in doublebuffering would actually need 64K.
On the other hand GX4000 (and CPC/PLUS) also suffered from having no proper VRAM at all.

[TotO]

Guys you're talking about an electronic masterpiece... NEC was far better than Amstrad concerning design.

Like I said, it's not NEC but Hudson who designed all those evolutive PC-Engine master pieces.

[emuola]

Like I said, it's not NEC but Hudson who designed all those evolutive PC-Engine master pieces.

Go Hudson, Go! 

[TotO]

Go Hudson, Go! 

May 18th, 1973 - March 1st, 2012.
R.I.P.

[emuola]

May 18th, 1973 - March 1st, 2012.
R.I.P.

Da*n, forgot about that One of the last "old giants".

[remax]

Da*n, forgot about that One of the last "old giants".

The ones that survived had less talent in innovating than in stealing ideas and marketing...

[MacDeath]

The ones that survived had less talent in innovating than in stealing ideas and marketing.

are you implying Apple was never cheap nor innovative ?
oh wait...


But hey, still NEC provided the production, and boy they were good at that.


Remember the NEC PC6xx1 used to have speccy specs, then CPC specs, then... CPC with better resolution specs ? (sort of)