What could have been for Dreamcast....
- Thread starter Crazyace
- Start date
- Status
First of all, I'd agree that Sega's problems weren't hardware related (disreagrding the difference in hype surrounding the DC and PS2 hardware).
The DC as released in 1998 was (IMO) a superb peice of hardware, the only change I might have made being leaving out the modem. Hard to say how this would have affected sales in the long run, but given the cost it added per unit (somewhere between $20 and $30 per unit at launch I think) it probably wasn't worth it. The 33K European modem wasn't particularly impressive in 1999 or 2000 either.
I'm inclined to agree with those that think a DC released in late 1999 would have looked like the DC we got, but faster and with probably a little more memory.
As for a DC released in March 2000, well, I don't think it would have looked anything like Naomi 2. For a start Elan wouldn't have been ready in time, and just as importantly without the DC we saw in 1998 there probably wouldn't have been Naomi 2 in 2000/2001 to base a home system off. Look at it this way - Naomi and DC were designed at the same time to basically use the same hardware in two complimentary markets. Without DC there would have been no Naomi, without the 1998 DC and Naomi there would be no Naomi 2. Therefore, you couldn't base a 2000 DC on hardware derived from an earlier variant of itself that never existed.
I think the best you could have hoped for with a March 2000 DC would have been something like...
- 133mhz Kyro based chip with the same kind of modifications the PVR2DC got, with 16MBs of 128bit 133mhz SDRAM
- dual SH4 cpus running at 266 mhz with 32 MBs of 128bit 133 mhz SDRAM.
Maybe you could push the clocks a little higher to a 150/300 combination like the PS2. Regardless, I think this would have made for a very competitive peice of hardware, and it seems a fairly realistic spec to me, but then again I have no idea how much this would have cost and wouldn't know where to look to find out.
Would it have been possible to put two SH4s on the same chip by March 2000?
The DC as released in 1998 was (IMO) a superb peice of hardware, the only change I might have made being leaving out the modem. Hard to say how this would have affected sales in the long run, but given the cost it added per unit (somewhere between $20 and $30 per unit at launch I think) it probably wasn't worth it. The 33K European modem wasn't particularly impressive in 1999 or 2000 either.
I'm inclined to agree with those that think a DC released in late 1999 would have looked like the DC we got, but faster and with probably a little more memory.
As for a DC released in March 2000, well, I don't think it would have looked anything like Naomi 2. For a start Elan wouldn't have been ready in time, and just as importantly without the DC we saw in 1998 there probably wouldn't have been Naomi 2 in 2000/2001 to base a home system off. Look at it this way - Naomi and DC were designed at the same time to basically use the same hardware in two complimentary markets. Without DC there would have been no Naomi, without the 1998 DC and Naomi there would be no Naomi 2. Therefore, you couldn't base a 2000 DC on hardware derived from an earlier variant of itself that never existed.
I think the best you could have hoped for with a March 2000 DC would have been something like...
- 133mhz Kyro based chip with the same kind of modifications the PVR2DC got, with 16MBs of 128bit 133mhz SDRAM
- dual SH4 cpus running at 266 mhz with 32 MBs of 128bit 133 mhz SDRAM.
Maybe you could push the clocks a little higher to a 150/300 combination like the PS2. Regardless, I think this would have made for a very competitive peice of hardware, and it seems a fairly realistic spec to me, but then again I have no idea how much this would have cost and wouldn't know where to look to find out.
Would it have been possible to put two SH4s on the same chip by March 2000?
A hikaru powered system would have been interesting to see....
Why so much memory waste? IIRC, the voodoo5 only duplicated the framebuffer for both chips, but most of the memory was used as one giant storage which both chips shared without duplicating data. And why a large memory pool for the elan chip, can't it share?
I think any of these proposed systems would have failed against ps2. In the short run, the power of the systems was irrelevent due to sony's marketting muscle and better name, and in the long run I still think the ps2 would have shown itself as more powerful than any of the systems...heck, some ps2 games often looked better than all similar titles on gamecube and xbox...ps2 just seems to throw polys and particles around much better.
Kyro 1 would have been interesting though, it compares to the original geforce very well, and probably the average game would have looked about the same as the average gamecube game, but I bet the best gamecube games would look much better.
A kyro 2 I think wouldn't fair much better though, but maybe it's average game may have looked as good as the average xbox game, but I don't think its best would look as good as the best on gamecube or xbox.
As far as I know, not one dc game supports surround sound, and even the stereo sound isn't on par with the latest systems.(at least it doesn't sound as good to me, but it may be the budgets dc games were made on) Pretty big since even psx and n64 had a fairly large selection of games with dolby pro logic.
Sega did have a broadband adapter, but unlike the newer systems where nearly every game is broadband only nearly ever sega game was modem only. I think only four games supported the bba, quake 3, unreal tournament, pod speedzone, and ...that shooting game by sega...Outtrigger!
Megadrive1988- Your ideal dreamcast sounds a lot like a gamecube to me without any sacrifices. Basically a gamecube with more ram, and I guess it's original clock speed of 200 mhz gpu.
BTW, I thought the initial high price of the ps2 was only because sony was about to miss the launch deadline and had to overnight or 2day ship 500,000 ps2s or something.
Why so much memory waste? IIRC, the voodoo5 only duplicated the framebuffer for both chips, but most of the memory was used as one giant storage which both chips shared without duplicating data. And why a large memory pool for the elan chip, can't it share?
I think any of these proposed systems would have failed against ps2. In the short run, the power of the systems was irrelevent due to sony's marketting muscle and better name, and in the long run I still think the ps2 would have shown itself as more powerful than any of the systems...heck, some ps2 games often looked better than all similar titles on gamecube and xbox...ps2 just seems to throw polys and particles around much better.
Kyro 1 would have been interesting though, it compares to the original geforce very well, and probably the average game would have looked about the same as the average gamecube game, but I bet the best gamecube games would look much better.
A kyro 2 I think wouldn't fair much better though, but maybe it's average game may have looked as good as the average xbox game, but I don't think its best would look as good as the best on gamecube or xbox.
As far as I know, not one dc game supports surround sound, and even the stereo sound isn't on par with the latest systems.(at least it doesn't sound as good to me, but it may be the budgets dc games were made on) Pretty big since even psx and n64 had a fairly large selection of games with dolby pro logic.
Sega did have a broadband adapter, but unlike the newer systems where nearly every game is broadband only nearly ever sega game was modem only. I think only four games supported the bba, quake 3, unreal tournament, pod speedzone, and ...that shooting game by sega...Outtrigger!
Megadrive1988- Your ideal dreamcast sounds a lot like a gamecube to me without any sacrifices. Basically a gamecube with more ram, and I guess it's original clock speed of 200 mhz gpu.
BTW, I thought the initial high price of the ps2 was only because sony was about to miss the launch deadline and had to overnight or 2day ship 500,000 ps2s or something.
Evil_Cloud
Veteran
Deepak said:
What I can remember, Conker's Bad Fur Day... But I am sure there are more.
Perfect Dark did, Syphon Filter 2 did(and presumebly 3, and maybe 1), and Rogue Squadron, and probably Battle for Naboo did.
Those are the only ones I can recall offhand, and more n64 games had it than psx games, but I'm sure there are more.
I think 1 SNES game had dolby pro logic audio too.
Those are the only ones I can recall offhand, and more n64 games had it than psx games, but I'm sure there are more.
I think 1 SNES game had dolby pro logic audio too.
Probably this has been asked much but can DC handle GT4 or MGS2/3?
I ask this because the famous Fishie (see http://www.beyond3d.com/forum/viewtopic.php?t=12782&start=0) claims that it could easily handle them. He says that GT4 and MGS2/3 are low poly games but made look so good because of nifty artesy effects and filters and long development time but even the DC could handle it he says... GT4 has 50 to 60 thousand polygons /frame at 60 fps (he got this from a developer... ?) what makes it 3,6 million polygons/second in total. So he concludes that DC could easily handle GT4, because Le Mans was 5 million polys/sec
But aren't there other factors in place? AI, car handling, special effects (like the rain effect in MGS2),... and the DC cpu is only 200 Mhz and it has only 16 Mb RAM (but 8 MB videoram) I don't know much about this technical stuff but there was a discussion on my forum with fishie I hope that people wo have experience with development on the DC can answer these questions, or fishie wouldn't believe it :?
I ask this because the famous Fishie (see http://www.beyond3d.com/forum/viewtopic.php?t=12782&start=0) claims that it could easily handle them. He says that GT4 and MGS2/3 are low poly games but made look so good because of nifty artesy effects and filters and long development time but even the DC could handle it he says... GT4 has 50 to 60 thousand polygons /frame at 60 fps (he got this from a developer... ?) what makes it 3,6 million polygons/second in total. So he concludes that DC could easily handle GT4, because Le Mans was 5 million polys/sec
But aren't there other factors in place? AI, car handling, special effects (like the rain effect in MGS2),... and the DC cpu is only 200 Mhz and it has only 16 Mb RAM (but 8 MB videoram) I don't know much about this technical stuff but there was a discussion on my forum with fishie
I hope that people wo have experience with development on the DC can answer these questions, or fishie wouldn't believe it :?Evil_Cloud
Veteran
I would ask it more like this: suggesting a developer had four years, all the funds in the world and a development team consisting of 200 persons, and had to develop GT4 or MGS2/3 for Dreamcast that is as good as the PlayStation 2 version now, would it be possible?
No, the Dreamcast wouldn't be able to replicate GT4 as nice as the PS2 version is. Test Drive: Le Mans pretty much maxed out the Dreamcast hardware. The team behind that game did things on the Dreamcast such as support a plethora of features that few developers would support one on their own. I know of very little other games that use anisotropic filterin gon the DC.
LE Mans was an excellent game on the DC. I wish there would be a sequel for a next gen system.
LE Mans was an excellent game on the DC. I wish there would be a sequel for a next gen system.
Evil_Cloud
Veteran
Lazy8s said:
Not identical looking, but as good looking (in a different way, using the various advantages of the system).
Evil_Cloud said:
Therefore, it won't be the same...
Infogrames's Melbourne House and their Le Mans 24 did indeed offer the best looking racer on the DC. So, if someone want to imagine a DC GT4, just look at Le Mans 24. Also, Melbourne House has a lot of staff ~100 persons that might help with the "a development team consisting of 200 persons".(The biggest part of any team are always dedicated to content creation(arts, game design,...))
BTW the engine "could, in theory" push 5Mpps, just like their PS2 Grand Prix Challenge engine , if you ask them, could push roughly 18Mpps...
I think using SEGA's F355 engine, the DC would come close to GT3 if the same development resources were allowed. You have to remember that TDLM's engine was designed to allow for what 20+ cars on the screen? If you could narrow that down to six cars like F355 or even four due to memory restrictions, then it may be able to come close, maybe within 90%?
- Status
