Original XBOX designed with AMD

[Madpacket]

I wonder what GPU they would have went with if they stuck with AMD. Perhaps this would have made a larger difference than the chosen CPU's. Maybe they would have stuck with nVidia or perhaps something better from ATI?

 

[NTMBK]

I wonder what GPU they would have went with if they stuck with AMD. Perhaps this would have made a larger difference than the chosen CPU's. Maybe they would have stuck with nVidia or perhaps something better from ATI?

The nVidia decision was made separately. The close runner up was actually Gigapixel, not ATi. Gigapixel had literally made the deal with Microsoft, when their CEO foolishly phoned up Jen Hsun Huang (NVidia CEO) to gloat. This got JHH fired up enough to come back to Microsoft with an even more competitive offer and win the contract out from under them at the last minute. (Gigapixel ended up going out of business, and being bought up by 3dfx.)

 

[Madpacket]

The nVidia decision was made separately. The close runner up was actually Gigapixel, not ATi. Gigapixel had literally made the deal with Microsoft, when their CEO foolishly phoned up Jen Hsun Huang (NVidia CEO) to gloat. This got JHH fired up enough to come back to Microsoft with an even more competitive offer and win the contract out from under them at the last minute. (Gigapixel ended up going out of business, and being bought up by 3dfx.)

Whoa, been a while since I heard of Gigapixel. Probably a good decision in hindsight by MS.

FWIW I still regularly use my crystal XBOX and two S controllers which has been modded with a 250GB drive and run all sorts of arcade emulators, has approx 3000 games installed.

Most games before the year 2000 run really well on it. It'll be a sad day if my XBOX ever dies.

 

[therealnickdanger]

Tell us how to make an emulator for the original Xbox and ill be happy.

QFT

How is this not a thing?

 

[nismotigerwvu]

What relevance does your second link have?

You admit you compare with lower FSB models. And the Anandtech article specificly saying the FSB is the limit. A limit that is offset in the Xbox CPU using a 133Mhz FSB.

Do you really think the change from a 100mhz fsb to a 133mhz fsb was going to close the gap? The 100mhz fsb 800mhz Celeron was slower by a measurable amount than 600mhz Duron. You'd need better than 1:1 scaling on the fsb (spoiler alert, it doesn't come close) to just BREAK EVEN. That's also basing this off the assumption that it would just be an off the shelf Duron and considering that Intel went ahead with a custom sku, it wouldn't be a huge leap for AMD to have done something similar (perhaps futzed with the cache configuration or rolled in some other Athlon specific aspects). Take a look right here http://www.anandtech.com/show/440/10 . Crunching the numbers, you'll see the 500mhz part (66mhz fsb) gains 21.7% in fps moving up to 677mhz (30% clock increase) with a fsb increase to 133mhz (67.3% increase). With the 550mhz model we see a similar trend, 67.3% increase in FSB, 28.5% increase in clocks, 24.9% increase in fps. It does appear that the higher clocked models do benefit more from the fsb increase, but not nearly enough to close the gap. The elephant in the room here is that this is beyond a best case scenario as this is also overclocking much more, most importantly the AGP bus (the PCI bus and system ram should also see increases). If it can't scale enough here, it never will.

 

[nismotigerwvu]

QFT

How is this not a thing?

https://www.youtube.com/watch?v=bUvGPcB4VO8

Well, it isn't a thing yet, but progress is being made

 

[soccerballtux]

Gates was probably concerned about AMD's ability to supply enough cpus.

volume would have been so low it would have been completely fine. He and Intel rubbed eachothers backs

 

[soccerballtux]

Yeah, the Xbox was far and away the most powerful piece of hardware that generation CPU and GPU wise. Granted it came over 3 years after the Dreamcast and a year and a half after the PS2 so this isn't exactly unexpected. However, the distance it placed between itself and the rest of the pack (even the Gamecube which launched in a similar time frame) is remarkable. Sadly, most titles were just developed to the least common denominator, so most of this muscle went unflexed outside a few notable examples.

and they picked up on that one and made the Xbox360

and the XboxOne in particular.

I'm ok with the 360, that was a really well designed piece of hardware architecturally speaking but the XB1...no...

 

[SPBHM]

Whoa, been a while since I heard of Gigapixel. Probably a good decision in hindsight by MS.

FWIW I still regularly use my crystal XBOX and two S controllers which has been modded with a 250GB drive and run all sorts of arcade emulators, has approx 3000 games installed.

Most games before the year 2000 run really well on it. It'll be a sad day if my XBOX ever dies.

I sold mine in 2007 I think it was pretty good, hacked with a 120GB HD installed, great for 16 bit systems emulation, great for watching divx/xvid videos with XBMC, great video quality with component out up to 1080i/720P, and also the native games were quite good... it made miracles, running hl2/doom3 with 64MB of memory, obviously they didn't look as good as the PC but looked impossible for the competition (GC/PS2)

anyway, I don't think they would go with ATI and Nvidia gave them a very good chipset/IGP, like they used to do for the socket A platform (nforce and nforce 2)

and again the Xbox CPU was not a Celeron, it was faster than Celeron at the same clock, slower than P3 (all Coppermine)

 

[nismotigerwvu]

and they picked up on that one and made the Xbox360

and the XboxOne in particular.

I'm ok with the 360, that was a really well designed piece of hardware architecturally speaking but the XB1...no...

Yeah, the 360 was a pretty nice piece of hardware. The GPU was a really forward thinking design and served MS quite well. They may have been better off with different CPU design though as Xenon was a bit underwhelming.Granted MS was predicting the multithreaded performance would make up for the lesser single threaded throughput, a more potent standard dual core (along the lines of a Toledo core Athlon X2, a massaged PowerPC 970MP or even some sort of Power5 derivative. Honestly I'm not really all too knowledgeable on the POWER side of things but you see what I'm getting at here) would have eliminated a bottleneck.

 

[soccerballtux]

Yeah, the 360 was a pretty nice piece of hardware. The GPU was a really forward thinking design and served MS quite well. They may have been better off with different CPU design though as Xenon was a bit underwhelming.Granted MS was predicting the multithreaded performance would make up for the lesser single threaded throughput, a more potent standard dual core (along the lines of a Toledo core Athlon X2, a massaged PowerPC 970MP or even some sort of Power5 derivative. Honestly I'm not really all too knowledgeable on the POWER side of things but you see what I'm getting at here) would have eliminated a bottleneck.

what bottleneck? It performed approximately equivalently to a Pentium 4 HT at 3.2ghz. or maybe it was Pentium-D. Whatever it was, it was nothing to sneeze at

 

[NTMBK]

volume would have been so low it would have been completely fine. He and Intel rubbed eachothers backs

Microsoft weren't aiming for low volume when they were planning the XBox. They wanted to sell Playstation numbers.

Intel came back with a cheaper price and got the contract, there's no conspiracy.

 

[nismotigerwvu]

what bottleneck? It performed approximately equivalently to a Pentium 4 HT at 3.2ghz. or maybe it was Pentium-D. Whatever it was, it was nothing to sneeze at

The Xenon was/is generally considered underwhelming at best and a "dog" at worst. Anandtech (written by Anand himself) ran an article covering this "Microsoft's Xbox 360 & Sony's PlayStation 3 - Examples of Poor CPU Performance", but had to yank it due to MS and/or Sony threatening to make their relationship a bit less cooperative. You can find reposts everywhere since nothing is ever truly "gone" from the internet. Here's one example http://gprime.net/board/showthread.php?p=89258#post89258

Important snips:
"While Microsoft and Sony have been childishly playing this flops-war, comparing the 1 TFLOPs processing power of the Xenon CPU to the 2 TFLOPs processing power of the Cell, the real-world performance war has already been lost."

"Right now, from what we’ve heard, the real-world performance of the Xenon CPU is about twice that of the 733MHz processor in the first Xbox. Considering that this CPU is supposed to power the Xbox 360 for the next 4 - 5 years, it’s nothing short of disappointing. To put it in perspective, floating point multiplies are apparently 1/3 as fast on Xenon as on a Pentium 4."

"The reason for the poor performance? The very narrow 2-issue in-order core also happens to be very deeply pipelined, apparently with a branch predictor that’s not the best in the business. In the end, you get what you pay for, and with such a small core, it’s no surprise that performance isn’t anywhere near the Athlon 64 or Pentium 4 class."

"The most ironic bit of it all is that according to developers, if either manufacturer had decided to use an Athlon 64 or a Pentium D in their next-gen console, they would be significantly ahead of the competition in terms of CPU performance.

"While the developers we've spoken to agree that heavily multithreaded game engines are the future, that future won't really take form for another 3 - 5 years. Even Microsoft admitted to us that all developers are focusing on having, at most, one or two threads of execution for the game engine itself - not the four or six threads that the Xbox 360 was designed for. Even when games become more aggressive with their multithreading, targeting 2 - 4 threads, most of the work will still be done in a single thread. It won't be until the next step in multithreaded architectures where that single thread gets broken down even further, and by that time we'll be talking about Xbox 720 and PlayStation 4. In the end, the more multithreaded nature of these new console CPUs doesn't help paint much of a brighter performance picture - multithreaded or not, game developers are not pleased with the performance of these CPUs."

"Another way to look at this comparison of flops is to look at integer add latencies on the Pentium 4 vs. the Athlon 64. The Pentium 4 has two double pumped ALUs, each capable of performing two add operations per clock, that's a total of 4 add operations per clock; so we could say that a 3.8GHz Pentium 4 can perform 15.2 billion operations per second. The Athlon 64 has three ALUs each capable of executing an add every clock; so a 2.8GHz Athlon 64 can perform 8.4 billion operations per second. By this silly console marketing logic, the Pentium 4 would be almost twice as fast as the Athlon 64, and a multi-core Pentium 4 would be faster than a multi-core Athlon 64. Any AnandTech reader should know that's hardly the case. No code is composed entirely of add instructions, and even if it were, eventually the Pentium 4 and Athlon 64 will have to go out to main memory for data, and when they do, the Athlon 64 has a much lower latency access to memory than the P4. In the end, despite what these horribly concocted numbers may lead you to believe, they say absolutely nothing about performance. The exact same situation exists with the CPUs of the next-generation consoles; don't fall for it."

"And that's what we have here today, with the Xbox 360 and PlayStation 3. Both consoles are marketed to be much more powerful than they actually are, and from talking to numerous game developers it seems that the real world performance of these platforms isn't anywhere near what it was supposed to be."

 

[NTMBK]

The 360 and PS3 CPUs are so frustrating, because they made bad design choices because of politics.

The PS3's PPU was originally intended to have out-of-order execution, and no SMT. But because one of the architects wanted to "make his mark" on the design he pushed through ditching OoO and adding SMT. Then this same CPU design formed the basis of the Xenon, too.

 

[Tuna-Fish]

what bottleneck? It performed approximately equivalently to a Pentium 4 HT at 3.2ghz. or maybe it was Pentium-D.

Only on very specific synthetic loads. On actual real code, the Xenon was a total disappointment. The cpu had very good straight-line performance for very clean code, but also had ridiculous amount of glass jaws that made it stall for a very long time, and that you could not in practice avoid.

Years after launch, after tons of adapting the compilers used to be better for the CPU, MS considered 0.2 IPC for each thread to be roughly what the cpu was capable in normal compiled c++.

 

[Hi-Fi Man]

Xenon performance was indeed around a 3.2GHz Pentium D (after all optimizations are factored in) but remember the Pentium D was horrible in game performance due to terrible FP performance and a long pipeline.

Just look at game requirements for good console ports around 2006-2008, they usually recommended a Pentium D ~3.2GHz and once console developers got to grips the Xenon the requirements then jumped to a Core 2 Duo ~1.83GHz or Athlon 64 X2 ~2.4GHz.

 

[SPBHM]

The 360 and PS3 CPUs are so frustrating, because they made bad design choices because of politics.

The PS3's PPU was originally intended to have out-of-order execution, and no SMT. But because one of the architects wanted to "make his mark" on the design he pushed through ditching OoO and adding SMT. Then this same CPU design formed the basis of the Xenon, too.

I actually think the 360 was quite impressive and well balanced, it would have been better if HD was standard at launch (360 core had no HD only memory cards for save games lol), also the failure of HD-DVD and the use of DVD, it was not a big deal, but a clear disadvantage...

oh and, I know the 360 originally was intended to have only 256MB, but even 512MB, considering how long it lasted I think was a bigger deal than the CPU and GPU limitations, the CPU seems to have delivered and caused less optimization problems than ram in the end?

 

[NTMBK]

I actually think the 360 was quite impressive and well balanced, it would have been better if HD was standard at launch (360 core had no HD only memory cards for save games lol), also the failure of HD-DVD and the use of DVD, it was not a big deal, but a clear disadvantage...

oh and, I know the 360 originally was intended to have only 256MB, but even 512MB, considering how long it lasted I think was a bigger deal than the CPU and GPU limitations, the CPU seems to have delivered and caused less optimization problems than ram in the end?

Yeah, the RAM limit must have been a pain to develop for, but at least it was unified (unlike the PS3).

 

[Genx87]

The nVidia decision was made separately. The close runner up was actually Gigapixel, not ATi. Gigapixel had literally made the deal with Microsoft, when their CEO foolishly phoned up Jen Hsun Huang (NVidia CEO) to gloat. This got JHH fired up enough to come back to Microsoft with an even more competitive offer and win the contract out from under them at the last minute. (Gigapixel ended up going out of business, and being bought up by 3dfx.)

Heh I never heard this story. That is very amusing. I wonder of JHH did this when Nvidia eventually bought 3DFX.

https://www.youtube.com/watch?v=UMdmWjtDRPE

 

[SlowSpyder]

The original Xbox was rumored to come with an Athlon. I remember stories on rumor sites saying Intel was swooping in at any cost to get their CPU's in the machine, and they did. At the time AMD had better technology and I think Intel felt this would help keep their hardware in the spotlight as the choice of gamers.

 

[Blitzvogel]

Xenon performance was indeed around a 3.2GHz Pentium D (after all optimizations are factored in) but remember the Pentium D was horrible in game performance due to terrible FP performance and a long pipeline.

Just look at game requirements for good console ports around 2006-2008, they usually recommended a Pentium D ~3.2GHz and once console developers got to grips the Xenon the requirements then jumped to a Core 2 Duo ~1.83GHz or Athlon 64 X2 ~2.4GHz.

As weak as people like to say Xenon and Cell are in general performance, console developers pulled some crazy shit out of them. They are GFLOPS monsters afterall. I simply don't believe the "twice as powerful as Xbox CPU" in general IPC claim. There are Youtube vids of dual Pentium III 1400 MHz systems running modern multiplatform games, with interesting performance metrics. The only problem is that they are not comparable tests to consoles since the guy who did them performed the tests at full HD with a Radeon 4670. Even with all the graphics settings completely minimized except for textures (he usually leaves them on medium or high), it's not a comparable environment from the perspective of aiming to recreate the exact console experience. All those other graphics processes being minimalized does reduce draw calls I assume. However, it could be less of a processing issue, and more of a bus communication limitation as well. The vids are more just to prove the dual P3's prowess in comparison to modern silicon, so more research needs to be done to get a real perspective vs the past console generation.

 

[dark zero]

Well... Seems that how Xbox born is interesting and with surprises... How a Intel and nVIDIA combo didn't work well after all. And seems that this is the future...
Maybe it's time to MS to permanently leave the console división and focus only on PC. With that, they win in everything.

 

[Hi-Fi Man]

As weak as people like to say Xenon and Cell are in general performance, console developers pulled some crazy shit out of them. They are GFLOPS monsters afterall. I simply don't believe the "twice as powerful as Xbox CPU" in general IPC claim. There are Youtube vids of dual Pentium III 1400 MHz systems running modern multiplatform games, with interesting performance metrics. The only problem is that they are not comparable tests to consoles since the guy who did them performed the tests at full HD with a Radeon 4670. Even with all the graphics settings completely minimized except for textures (he usually leaves them on medium or high), it's not a comparable environment from the perspective of aiming to recreate the exact console experience. All those other graphics processes being minimalized does reduce draw calls I assume. However, it could be less of a processing issue, and more of a bus communication limitation as well. The vids are more just to prove the dual P3's prowess in comparison to modern silicon, so more research needs to be done to get a real perspective vs the past console generation.

Wow, I just looked up those P III videos and it's just crazy seeing people running those things with radeon 4670s and here I thought rebuilding my socket 939 gear was a bit of a waste!

It's pretty obvious from those videos that the CPU is bottlenecking hard even at lowest settings making xbox 360 performance and quality levels unavailable. Xenon (even Pentium D 805) is significantly faster than dual Pentium IIIs @ 1.4GHz. Yes Xenon has anemic cores but it has VMX-128 (basically Altivec with double the registers) which is comparable to SSE2. The Pentium III only has MMX and SSE.