Odyssey 3 Why didn't Philips re-release the Videopac G7400/O3 in response to the NES?

[Prosystemsearch]

The Atari 7800 got that treatment in response the the success of the NES, but not the Odyssey 3. WHY!? Philips wasn't in that big of a hampering position AFAIK.

 

[VPB---]

Currently, Philips does not even remember that they made video games in the 80's.

[Prosystemsearch]

Currently, Philips does not even remember that they made video games in the 80's.

 

And the latter half of the 1970s. Also, how do you know that they don't remember as a company, or do know, but refuse to admit it?

Edited January 14, 2017 by Prosystemsearch

[LS_Dracon]

G7400 have ZX-Spectrum-like graphical capabilities, would not be a strong competitor to the NES.

[ubersaurus]

Nothing about the Odyssey 3 makes me think it would have had a snowball's chance in hell in the mid to late 80s.

[Prosystemsearch]

G7400 have ZX-Spectrum-like graphical capabilities, would not be a strong competitor to the NES.

>ZX spectrum-like capabilities graphically.

Uhh, I don't think that is 100% true. Yeah it certainly doesn't measure up to the 8 bit powerhouse that is the master system technologically, but it was released in Europe around the same time the original Famicom was launched over in Japan, Plus, Philips might have also released 2-button joystick controllers and corresponding games had they gave it a chance in practice. Yeah, probably not the 300-baud modem as part of the re-release, but maybe whey would have added a special pause button as well as built-in voice synthesis for either the would-be new launch or at least for the later model called the "command center" for North American Odyssey 3.

[VPB---]

You're right. And I love to play with the Command Center.

Unfortunately Philips has always wanted to innovate in different systems but they sometimes stopped very quickly leaving the consumer in distress.

[Prosystemsearch]

Hmmm... I wonder how many G7400s got bought by consumers during its short official production run?

[Prosystemsearch]

Also, keep in mind that because the Videopac G7400 was released in a very bad time for video games in the west(especially in the Americas), it never got much time for programmers to show its full capabilities, and that is just it without the even more rare computer add-on.

Edited January 17, 2017 by Prosystemsearch

[alex_79]

I don't think the Videopac+/Odyssey^3 would have had any chance against the NES. The chipset used for the "enhanced" graphics of the videopac+ wasn't even specifically designed for video games. It was an off-the-shelf part designed for Teletext-like displays: 40x24 text, each character being 8x10 pixels, 8 colors x 2 levels of brightness.

To denote its intended destination, the chipset supported things like scrol down/up (entire rows of text, not in single pixel increments), zoom (by displaying the rows of characters at double height), blinking, reverse video, but didn't have features more useful for games like, for example, sprites that could be moved freely, horizontal scrolling (not even 1 character at a time), mirroring of the graphics, etc.

Even advanced controllers with extra buttons weren't considered for the console: the controller ports had no free inputs to be used for extra buttons, nor power or output lines to allow extra electronics in the controller themselves, and no analog inputs. There are 6 pins only: 5 inputs lines (for the 4 joystick directions and button) and ground, just like on the original Videopac/Odysssey^2.

As a comparison, each of the Atari 2600 controller ports has a power pin, 4 programmable bi-directional digital lines, 1 digital input, and 2 analog inputs, which offer a lot of possibilities for different controllers or other peripherals.

[Prosystemsearch]

I don't think the Videopac+/Odyssey^3 would have had any chance against the NES. The chipset used for the "enhanced" graphics of the videopac+ wasn't even specifically designed for video games. It was an off-the-shelf part designed for Teletext-like displays: 40x24 text, each character being 8x10 pixels, 8 colors x 2 levels of brightness.

 

To denote its intended destination, the chipset supported things like scrol down/up (entire rows of text, not in single pixel increments), zoom (by displaying the rows of characters at double height), blinking, reverse video, but didn't have features more useful for games like, for example, sprites that could be moved freely, horizontal scrolling (not even 1 character at a time), mirroring of the graphics, etc.

 

Even advanced controllers with extra buttons weren't considered for the console: the controller ports had no free inputs to be used for extra buttons, nor power or output lines to allow extra electronics in the controller themselves, and no analog inputs. There are 6 pins only: 5 inputs lines (for the 4 joystick directions and button) and ground, just like on the original Videopac/Odysssey^2.

 

As a comparison, each of the Atari 2600 controller ports has a power pin, 4 programmable bi-directional digital lines, 1 digital input, and 2 analog inputs, which offer a lot of possibilities for different controllers or other peripherals.

 

What makes you so sure that stuff like horizontal scrolling and mirroring graphics weren't possible on even the Odyssey? Did you get to open one up and examine the circuits?

[alex_79]

The consoles are well documented. Datasheets of the chips and complete schematics are available.

Those features are not supported directly by hardware, which doesn't mean they can't be implemeneted in software. But that comes at the cost of increased cpu usage and resources (rom/ram), which could have been used for other aspects of the games instead.

[carlsson]

Philips at that time sold MSX computers, and eventually was the only (?) European brand to go ahead with MSX2 systems. While it doesn't fully answer your question, I think part of the answer lies there. They also had the VG-5000 computer which though seems more on par with the G7400 than both MSX and NES. A number of Videopac developers seem to have been recruited by Philips and others to port their games both to VG-5000 and later on MSX computers, believe it or not.

 

It strikes me that the Korean manufacturer Daewoo in 1985 released MSX based consoles under the name Zemmix. In a distant parallel universe, Philips may have done the same, release a MSX console which to some part had looked like a slightly upgraded ColecoVision or SG-1000, but cartridge games compatible with their computer line. Obviously that had in no way been Videopac compatible, but at least an initiative from Philips to hang onto console gaming.

 

While the Magnavox Odyssey was among the first to release video games in the early 70's, I think that by the mid 80's, Atari overall had a stronger ongoing tradition about consoles than Philips/Magnavox had at that time which also explains why Atari wanted one more go on the market once Nintendo and Sega had revitalized it. Actually Atari repackaged the 8-bit computer line as XEGS so at one point they had two different, somewhat aged video game consoles on the market, just like how Commodore made the infamous C64GS.

 

Sales figures as per Wikipedia, recently updated:

NES/Famicom: 62 million (19.5 million in Japan, 34 million in Americas, 8.5 million rest of the world)

Mark III/Master System: 18 million (1 million in Japan, 2 million in US, 8 million in Brazil and 7 million in Europe)

Atari 7800: 3.7 million

XEGS: 100,000

 

See also some of the other console manufacturers at that time, e.g. Casio PV-1000 (Oct 1983, with 13 games released including Pooyan, Super Cobra, Amidar, Dig Dug etc) and Epoch Super Cassette Vision (1984, with nearly 30 games released including Boulder Dash, Dragon Ball, Mappy, Miner 2049er, Pole Position II etc - this system was also sold in France by Yeno). The latter pales next to the NES, but gives the G7400 a run for the money. If both a known gaming brand like Atari, and some Japanese brands able to obtain decent licensed titles (though some a bit dated by 1985) were struggling to keep up with Nintendo and Sega, I find it likely that a re-released G7400 in 1985 or even worse 1986 had been a laughing stock more than something that would sell any units.

 

I seem to recall that surplus inventory of the G7000 (as well as Intellivision, 2600 etc) was very sold cheap by the time the NES and SMS were starting to gain a market share, but it is one thing to get rid of old stock and something else to introduce a "new" system with software development, manufacturing, marketing etc.

 

By the way, I find it funny that the CD-i is listed as the successor to the G7400. Well, it was the next video game console released by Philips, but with a break of 8 years it is bold to call it a successor, rather a fresh start.

Edited January 24, 2017 by carlsson

[carlsson]

As for how many G7400 were sold, I tried to look up some serial numbers. I understand that the G7000 serial numbers may begin with 71, and the G7400's I found have serials 753307 (possibly manufactured week 38, 1983) and 760318. That is 7011 units inbetween, if those are true serial numbers. Of course neither is said to be the first or last manufactured. The Digital Press FAQ says the latest Philips production date found is week 3, 1985, but perhaps that applies to games more than hardware.

[BillLoguidice]

The bottom line is is that the Odyssey2 was not a huge seller, and no one would have given an Odyssey3 system a second look. It was not a strong brand and would have been flushing money away.

 

Now, I have several Odyssey2 and VideoPac systems, including a Videopac+ G7401, and there's nothing about them in their entire library - even for the enhanced stuff - that screams, "I belong in a post-Crash videogame world." (and this is coming from someone who really likes those systems)

 

Now, if you said "Why didn't Mattel try to bring the planned Intellivision III or IV to market to compete with the NES?", there might have been something of an argument there, even if it's one just for fun. However, the reality is, outside of Atari, none of the pre-Crash console makers (at least the ones still solvent) were really in a position to compete with Nintendo, nor was it like Atari was doing gangbusters. Times, and the market, changed.

[SSG]

I kinda want one.

[LS_Dracon]

These are the upgrades on O3/G7400 hardware I think would be good to make it stand among competitors (ColecoVision, Intellivision and Atari5200)

I'm thinking on affordable technology at time, that would not be difficult to produce:

 

1 - Graphics

 

Background :

Another video mode using 8x8 tiles. 10x8 is not good because 10 is not a friendly number to work with, in binary. Division by 8 is cheap.

You could keep 10x8 as another video mode, for text only.

And a 2bpp video mode (like on 5200), this makes possible to use 3 colors for each 4x8 tile.

 

Would be cool 128 customizable characters instead 64, if it will not be so expensive to produce. You could use bit 7 to select between 2 banks of 64 chars.

 

Sprites :

8 16x16 mono colored sprites. CV and 5200 had four (on same scanline).

Currently, G7400 had none! You need to stick with 8x8 low resolution O2 movelable objects.

 

Colors :

Possibility to use all 16 colors, not choose between 8 light or 8 dark.

 

2 - Sound

 

A GE or YM sound chip similar to CV, Inty or Vectrex. Cheap and good for the time.

Currently G7400 uses the terrible O2 mono channel sound chip which is bad even for sound fx.

 

3 - Hardware

 

O2 backward compatibility, a faster processor (I think it is on G7400), keyborard from O3 proto/Command Center, not the membrane from O2/G7400.

I don't know about RAM, more is better but not to make it expensive.

Free the video bus to make changes on the fly, like on Atari systems, featuring WSYNC and a proper scanline counter.

Hardware smooth scrolling, 2 button controller etc...

 

This system wouldn't had chance against NES but would be interesting to port arcade games from the time.

Edited January 25, 2017 by LS_Dracon

[carlsson]

From what I understand when reading the specs on Wikipedia, the MCU is the same, an Intel 8048 clocked at 5.37 MHz in NTSC or 5.91 MHz in PAL? Or perhaps that is what you mean by "faster", that the G7400 is capable of running 10% faster than the Odyssey^2 and G7000? However the G7400 is said to have 6 kB RAM + 128 bytes, which I suppose is CPU and A/V RAM for backwards compatibility.

 

Otherwise your spec sounds very much like a Colecovision (Sega SG-1000, Sord M5, Casio PV-1000 etc) but with a 8048 instead of a Z80. It strikes me that the C7010 Chess module for the Videopac already contains a NSC800, which is a Z80 compatible CPU. Perhaps if the O^3 was a dual CPU system with built-in Z80 and the 8048 for backwards compatibility, plus the other improvements you list, it had been something and actually been directly comparable with the Colecovision. Then it all would be up to the manufacturer to obtain licensed titles or third party developer support.

 

The G7420 is a BASIC module with .. Z80 and an extra 16K RAM plus 16K ROM. It is another pointer that the G7400 needed a built-in Z80.

[BassGuitari]

When I saw the title to this topic, I knew it was gonna be a Prosystemsearch thread.

[LS_Dracon]

From what I understand when reading the specs on Wikipedia, the MCU is the same, an Intel 8048 clocked at 5.37 MHz in NTSC or 5.91 MHz in PAL? Or perhaps that is what you mean by "faster", that the G7400 is capable of running 10% faster than the Odyssey^2 and G7000? However the G7400 is said to have 6 kB RAM + 128 bytes, which I suppose is CPU and A/V RAM for backwards compatibility.

 

Otherwise your spec sounds very much like a Colecovision (Sega SG-1000, Sord M5, Casio PV-1000 etc) but with a 8048 instead of a Z80. It strikes me that the C7010 Chess module for the Videopac already contains a NSC800, which is a Z80 compatible CPU. Perhaps if the O^3 was a dual CPU system with built-in Z80 and the 8048 for backwards compatibility, plus the other improvements you list, it had been something and actually been directly comparable with the Colecovision. Then it all would be up to the manufacturer to obtain licensed titles or third party developer support.

 

The G7420 is a BASIC module with .. Z80 and an extra 16K RAM plus 16K ROM. It is another pointer that the G7400 needed a built-in Z80.

 

I don't know the exact number, but 8048 on O2 is not even 1 MHz. I think it is 0.8 MHz.

Yes, the system I described is pretty much a ColecoVision with twice of sprites and a 5200 2bpp background.

We are talking about a 1983 or 1984 spec system, so make sense a bit better hardware.

 

EDIT : A dual processor would make it too expensive and difficult to code. a 5MHz processor is fine, even for 8048.

 

EDT 2 : Let's count : 22.75 cycles each scanline x 262 (NTSC) = 5960.5 cycles per frame

5960.5 x 60 frames per second = 357.630 Hz, divide by 1000000 = 0.3 MHz. OMG!

Edited January 25, 2017 by LS_Dracon

[alex_79]

EDT 2 : Let's count : 22.75 cycles each scanline x 262 (NTSC) = 5960.5 cycles per frame

5960.5 x 60 frames per second = 357.630 Hz, divide by 1000000 = 0.3 MHz. OMG!

 

Yup. The clock (5.37MHz for NTSC / 5.91MHz for PAL) is divided by 3 inside the 8048 (1.79MHz/,1.97MHz) then again by 5 to get the "instruction clock" (0.36MHz/0.39MHz). Each instruction takes 1 or 2 of these instruction cycles.

 

(source: Intel MCS48 Family of Single Chip Microcomputers User's Manual)

Edited January 25, 2017 by alex_79

[alex_79]

As for the ram, even if there are effectively 3x 2K ram chips in the console, only 4k is accessible as two of them have one of the address line tied to ground. 2k chips were more common and cheaper than 1k ones at the time!

[LS_Dracon]

I knew the processor is slow but not that much!

I aways have problems coding my homebrews, VBlank is not enough to hold the game logics and is a pain try to reduce the instructions, sometimes, wasting more RAM, to make the code work faster.

 

As for Chess module, O2 CAN hold a chess game without module. Probably Philips used the module for 2 reasons :

 

1 - They have a good Z80 chess code around and

2 - They wanted a faster game than the Atari 2600 version.

 

Nanochess made an exelent 1kb chess for Atari 2600 which I was tempted to translate for O2, as I know how to code for both machines, just to prove my point, but man, I don't like Chess...

Maybe one day I try to get his authorization to do it.

Edited January 25, 2017 by LS_Dracon

[carlsson]

OMG! I read on Wikipedia that with a 11 MHz max crystal, one would get about 0.5 MIPS from the 8048. Would the Odyssey^2 then produce half of that? For comparison, the dry figures for a 4 MHz Z80 is 0.58 MIPS and 0.43 MIPS for a 1 MHz 6502. While it is stupid benchmark, it puts things into perspective.

 

Anyway, I saw that the G7400 was introduced in November/December 1983 at a price of 1590 Franc (1490 Franc in the review later in the magazine):

http://download.abandonware.org/magazines/Tilt/tilt_numero008/TILT%20-%20n%B008%20-%20novembre%20decembre%201983%20-%20page086%20et%20page087.jpg

 

A few price comparisons from the very same issue of Tilt magazine:

 

Interton/Hanimex VC-4000: 1100 Franc

Mattel Aquarius: 1200 Franc

Atari 2600: 1450 Franc

ColecoVision: 1790 Franc (RRP 2000 Franc)

Intellivision: 1890 Franc including 4 games (RRP 1950 Franc)

 

So in order to be competitive with the ColecoVision, Philips had perhaps up to 500 Franc (75 Euros?) more in the consumer end to spend on improvements to make a lasting console.

[CatPix]

Nice to see the prices comparisons. Pretty amazing to see that the VC-4000 still sold for 1100 Francs at the time - the minimum salary at the time was around 5 500/6 000 Francs, so if we compare by today's 1200 minimal wage, that's about 220 €; Compared to the price asked for an Atari 2600... That's steep