Maybe it dropped in price, but when I requested the materials and application form when it was first released, the Yaroze was $700. Of course, at the time, the PS1 was around $300...
Yes, you are. You know when they're needed. If you pull them off of the storage medium, they're needed - so you plop them in memory contiguously. You don't stick one texture at 0-5M, one at 417-419M, and one at 511-512M.
In practice, perhaps. I didn't deny that. You *could* however, actually do that in a unified memory structure but not in a shared memory configuration. Is this an advantage? Perhaps, perhaps not. You want 400 megs of textures? You got it. No difference from running 100 megs of textures.
And Microsoft still lost money on it for virtually the entire time they sold it. Why? Because it was poorly engineered. It was more expensive than it needed to be to produce the graphics it did.
MS brought a console to market, with games, in a single year. They also didn't have the advantage of hardware relationships and none of the hardware components were designed in-house. It wasn't so much that they designed it poorly, they did the best they could in the short amount of time they had because they had to use components which were already researched and tested. Compare this to PS2 which was in the making for at least 3 years while Sony tinkered with the EE to get it up and running. They own practically everything, so it's trivial to manufacture it on any process they want, even combining dies if they need to.
not the least of which is because 25% higher latency is *nothing*
I'd like to point out that's real-world measured latency, not absolute latency. RDRAM had a latency of around 11 clocks on first read, if I remember correctly. I'm not sure what DDR had at the time, but I'm sure it wasn't more than 4.
The GPU is well-designed for a shared-memory design - it offloads a lot of the bandwidth intensive tasks to a high-bandwidth local storage solution. But it is still strictly worse than if the design was 384MB of system RAM and 128MB of GPU RAM with a 10MB eDRAM solution, and some sort of GPU system RAM mapping. That would've been more complicated (and more expensive), however.
Remember, we're putting the CPU on the GPU memory bus and not the other way around. Sure, your solution is going to be faster. A ton faster? I don't think so. The reality of it though is that you'd really be choosing between this: shared memory bandwidth and 10M of EDRAM that will drop siginificantly in price or segragated memory busses that won't drop in price and no EDRAM. I'm guessing it's about equal once you factor 4xMSAA into the mix except for the fact the first solution will be cheaper in the long run.
It's not the absolute latency that matters - it's the clock latency.
Yes, that is correct, but they don't generally spec ram latency in ns, they spec it in clocks. Cas 3-4-4-8, isn't 3 ns. It's 3 clock cycles. I believe GDDR3 is 5-6-7-9 or something like that (who knows what's actually in the 360). So yeah, you're doubling the latency in some situations, but you're transferring data at a much higher rate to more than make up for it. RDRAM, btw, had a much higher latency than that when it first came out and DDR2 isn't all that great either, when compared to DDR1. So yeah, there's a penalty, but in real world applications, it's not going to be much more than 10-20% compared to equally clocked DDR1 (not that you could clock it that high).
A moderate latency solution can be fine for a design with only 1 thread and 1 GPU clamoring for bandwidth, but it can be much less efficient with 6 threads and 1 GPU as well.
Which is where great programmers will make the difference over good programmers. Take a look at GT4 and God of War for PS2. Do those games look great because the PS2 is a well balanced machine? No, they look good because the programmers know the hardware inside and out and can schedule data flow to it's maximum effect. I'd also say that the 360 is a better balanced machine than
Re:Too bad he's running the site off on 28.8 Kbps
on
Tinfoil Hat House
·
· Score: 1
Hey, I'm not arguing the nosey neighbor side, just correcting the facts. The basketball hoop is in the driveway now, because it was eventually moved there. If the story is factual, the basketball hoop was much closer to his bedroom window previously.
Re:Too bad he's running the site off on 28.8 Kbps
on
Tinfoil Hat House
·
· Score: 1
Go look at the pictures. You can see that the place he put the basketball net is in his driveway where people usually put such things. The horror! So his neighbor isn't supposed to play basketball in his driveway on a saturday morning? Maybe he should just stay in his house and take pictures of the neighbors through his blinds instead. I know which one of these guys I'd prefer living next to...and its not the little weasel with the camera.
If you had read the entire story, you would have found that the neighbor's basketball hoop had been vandelized by someone throwing bricks through the backboard (also damaging the writer's house). The neighbor then put the hoop by the street for the garbage men to pick up. Not knowing if it was trash or not, the garbage men never took it. The neighbor then proceeded to put it in his driveway (with a broken backboard) as shown in those pictures.
It'd be silly for it not to be. It would just increase the latencies for graphics reading, and you wouldn't be able to use the full system bandwidth.
Bzzzt! Wrong. It would be silly to not make *textures* contiguous or models or anything that's going to be read all at once, but in a game, you won't be able to line your textures all nice-and-neat in a row in the order they're needed. Would it be used like this in practice? Who knows, but it's quite irrevelent. The fact is, you *can* and that's the difference between unified memory and shared memory.
Of course it doesn't need the frame in eDRAM to work on it, but the fact that you basically can't store the frame in eDRAM should tell you that it's quite small. It's not a texture cache. At least, not for next-generation textures.
Before making statements like this, perhaps you should read what it is you're talking about first. 720p *can* fit entirely in the EDRAM with room to spare, but in practice, they'll most likely be splitting it up to make the process more efficient. BTW, I doubt seriously that many, if any at all, of the textures will be stored uncompressed. At 6:1 compression, even "modern" textures will fit quite nicely in cache. You won't be able to store them all, obviously, but more than enough to work on any particular chunk.
The Xbox was not a good design. It was a poor design that competed solely by brute force. The unified memory architecture was panned greatly when it was announced.
And yet, it's proven quite capable despite this.Take a look at Forza Motorsports, Halo 2 or any modern game on the XBox and compare it to computer games today. This "poor design" seems to be keeping up quite well. Perhaps not in resolution, but then you've said it yourself, that doesn't count.
Besides, it's important to remember that the Xbox 360 is a scale up from the Xbox. Latencies that wouldn't hurt the Xbox much will hurt the Xbox 360 much more.
You know, you seem awfully fixated on latencies. A few years ago, the gaming industry had a similar argument about this... RDRAM VS SDRAM. Despite having real world latencies 25% greater than SDRAM, RDRAM still managed to beat out SDRAM in real world games. And this is with a pitiful increase in bandwidth compared to the 3.5x increase in bandwidth the 360 is getting. I would mention that real world latencies take into effect the number of cycles VS the speed of the memory. I would imagine the real-world latencies (can't tell until their benchmarked) would be much less for the GDDR3 running on the 360 at 700Mhz VS DDR2 at 400Mhz running on a P4. All you have to do to see the real world performance of upgrading to lower latency memory on computers today to see the difference isn't all that great.
Yah, basically. Guess I'm just demanding.
.
Guess so... good thing the rest of the world doesn't seem to see it your way. Me I'll stick to a generation gap performance increase of 10x.
No, it's not. It's not even the same from a semantic definition. If you have a bundle of 4 oranges and 4 apples, and a bundle of 8 oranges and 8 apples, the second bundle isn't 4X more fruit because it's got 2X the oranges and 2X the apples. It's twice more fruit
I'd like to expand on this.
According to you, the GeForce 6800 Ultra is only about 4x more powerfull than a Voodoo 5-5500 and only 25% faster than the onboard video of the 925 chipset because the 6800 Ultra can only run Quake 3 at 400fps VS the Voodoo 5's 100fps and the 925's 300fps? Actually, it's not even that because this is only 4x the framerate, not 4x the polygons, 4x the texture resolution, 4x the whatever else you decide to ignore as a factor. Oh and I've completely ignored the fact the 6800 Ultra is running at 1280x1024 and not 1024x768 because that's not a factor either.
BTW, "a whole of it's parts" would be having a fruit basket with 8 fruit because you have 4 apples and 4 oranges (and a basket). "A sum of it's parts" would be having 4 apples and 4 oranges and adding them to get 8 fruit. The real term you are looking for is "The median of it's parts," or "it's only as strong as it's weakest link."
Which, by the way, isn't found by looking at the bandwidth and seeing that the 360 only has 3.5 times as much. The two systems use bandwidth completely differently and we really won't know for a couple of years if that's really the weak point of the system or not. Especially since 4xMSAA uses 0% of the main system bandwidth and the 10MB of EDRAM is on a seperate bus (caching texture, z-buffers, frame buffers and fragment programs).
Because if you want to have 128M of available memory for graphics-card related objects, you only have 512M-128M available for things that don't involve graphics. And if the GPU is working on something in that 128M, the CPU can't write there (at least, not without causing graphics corruption).
Who says the memory has to be contiguous? Load a texture, dump it in memory from disk and the GPU has instant access, with no additional writes to GPU memory. I've yet to see an onboard GPU in modern computing that didn't have your memory option in BIOS... not exactly on-the-fly re-distributing of memory.
I don't ignore it. Without it, the entire design would be laughable. The 10MB is critical, but it's not a texture cache: it's a framebuffer. At the Xbox 360's maximum resolution, the framebuffer will take up basically the entire 10MB.
Er... you didn't actually read the article I linked to did you? The GPU does not need to have the entire buffer in EDRAM to work on it. In fact, the ATI rep mentioned that 1080p was not a problem for the GPU, if MS decided to implement it. It can work on chunks of the screen at the time, buffering texture, z-buffer, fragment programs, whatever. You also seem to completely gloss over that the xbox 1 has exactly the same design minus the edram. And yet, this "laughable" design creates games that look better than Sony's PS2 with gobloads of memory bandwidth.
No, it's not. It's not even the same from a semantic definition. If you have a bundle of 4 oranges and 4 apples, and a bundle of 8 oranges and 8 apples, the second bundle isn't 4X more fruit because it's got 2X the oranges and 2X the apples. It's twice more fruit.
Well, isn't that a little obtuse. So you're saying that a machine that can easily push 6x the number of pixels at 720p isn't really 6x more powerful because thats only 720p vs 480p and 4x MSAA vs no AA. And just because this machine, that pushes 6x the pixels, can also quadruple the number of polygons at the same time, that doesn't count. In fact, if it were a real machine, it would be able to do 6x the polygons. Even though they would be untextured without shaders running. In fact, if you included shaders and textures and everything else, it could probably manage more than 4x the number of polies, but that doesn't really count either. All I have to say is that you have a strange sense of more powerful computing hardware.
That's the main memory bandwidth, not the graphics bandwidth. There are a few issues there - that's now split among 6 front ends, that only have 1M of shared L2 cache between all of them. The latency of GDDR3 is good for high-bandwidth, but not wonderful all-around - sharing it between 7 users will really start to lower its effective bandwidth quite heavily, because instead of streaming data, you're doing a whole lot of commanding.
True, and we saw this with RAMBUS on the PC not long ago... of course, we're talking about 3.5 times increase over bus speeds for today's modern processors and the P4 didn't suffer *that* much from high latentcy RAMBUS memory in real world applications. Heck, dual core P4's will be in a worse situation, bandwidth-wise, than the 3 core 360 processor.
Either that, or you're going to have to sacrifice on the amount of parallelization that you put in your code, which is probably more likely. I'd bet a lot that several of those front ends will just sit idling with a thread that, I dunno, checks for controller input or something.
In the first year or two, I'm sure you'll be right. After that? I'm not so sure.
Uh... the shared RAM means that the GPU *does* segragate a chunk of system RAM. That's what a unified memory architecture means. If you're saying that that chunk might not be contiguous, as it is with standard shared systems, yah, sure, but that's not that big a benefit. It's still shared bandwidth.
Maybe you have a differen't definition of segregated than I do. The CPU is in no way limited from writing/reading anywhere in system memory and neither is the GPU. How is that segregated?
Unified memory is strictly worse than separate GPU memory, because you can always get all of the "benefits" of unified memory by having something like an AGP GART on the GPU which allows the GPU to access main memory as well as its own local memory. The onboard memory then becomes a cache for the most often-used textures. Incidentally, you can guess how useful this feature actually is by looking at the number of AGP cards that actually *used* the GART to access main memory - as in, basically none. All graphics objects are heavily used. You want them close to the GPU.
GART sucked. Notice how both ATI and Nvidia are going back to this now that we have PCIe? Of course, main system bandwidth sucks (6.4GB/s), so it isn't all that usefull, yet.
But this also brings up the 10MB of EDRAM in the GPU you keep ignoring. Tech Report http://techreport.com/etc/2005q2/xbox360-gpu/index.x?pg=1 has an interesting article on the internals of the GPU. Not only is the 10MB used as a cache, but the circuitry surrounding the memory itself can do 4x MSAA for "free." In essence, the GPU isn't going to be going to main system RAM nearly as much as a typical PC Video card will. Not only that, there is no bandwidth required from the main system memory to do MSAA.
MS chose unified memory because it's far cheaper, and simpler to design. One bus, rather than two.
I was thinking dual-ported ram, as would be the straight-forward way of doing this. The GPU is also the memory controller, though, so it is a bit cheaper. BTW, you still need two busses, one from the CPU to the GPU and one from the GPU to the Memory, so it's not as huge a savings as you make it sound. You basically save the cost of needing two memory controllers instead of one.
Uh... about 4X more powerful. That's kinda my point. You're stacking the improvements, but a game is the whole of its parts, not the sum of them. If there are more polygons, but the textures still look crappy, it wouldn't necessarily be better.
er... the whole of its parts and the sum of its parts is, well, the same thing... In any case, I know wha
I'm sure the "15X" numbers were generated by averaging the peak FLOPs, bandwidths, etc. for the Xbox vs. the Xbox 360, but most of the boost comes from the CPU, which is "80X!!" more powerful than the original (115.2 GFLOP vs 1.5 GFLOP).
Yeah, you're right on this one... I just read a few places that showed MS was claiming 70 GigaFLOPS for the Xbox 1... wow, I'm baffled:-)
The important numbers, in my mind, are 22.4 GB/s - bandwidth to main memory - which is 3.5X the Xbox's bandwidth, but now it's shared between 6 threads and a GPU rather than 1 thread and a GPU. You can't get 15X performance - can't - with only 3.5X the bandwidth (and equivalent latency).
Wow, 3.5x the bandwidth? That's pretty big. More than I thought, really. That's much more than the difference in bandwidth between a GeForce 6800 and a GeForce 3.
This next-gen leap is about equivalent to each of the previous ones.
I agree. If you were to tell me that the PS2 was 10x more powerful than a PS1, I'd say you might be lowballing it a little. Texture Perspective Correction alone is a huge just by itself. In that respect the PS2 is infinitely more powerful because the PS1 couldn't do that basic function at all. I wouldn't say the same for the Gamecube VS the N64, but then the N64 was quite a bit more powerful than a PS1.
Last I heard, Xbox 360's peak polygon fill rate is 500M/s, not 6G/s, compared to the Xbox's 116.5M/s. So, uh, no, it can't. It can dump a lot more textures on those polygons - about four times more - but in terms of raw polygons, no, it can't.
You're right there... Not sure where the 6 billion pollies came from as I can't find the source I was looking at... Just goes to show you how reliable Google is at times:-P
But, let's go back to your numbers. If the Xbox 360 could do 400 M/s and the xBox 1 could do 100 M/s (just rounding to make things easier), but could texture them the same, would that make the Xbox 360 4x more powerful? Now add in 4x as many textures... how do you measure that?
It's not my definition of "more" powerful. It's my definition of 15X more powerful. I never said it wasn't more powerful - it's just not "15X" more powerful. That's just silly. It can't run at 15X higher resolution, for instance. It can't have textures with 15X the detail of the original.
In my opinion, you don't have to do 15x the resolution to be 15x more powerful. As I see it, 720p is 3x the pixel count of 480p. Add 4x the polly count (have you seen the alpha games like Test Drive?) and 4x the texture quality, where does that leave you? And that's just basic graphics that doesn't even take into account what you can do with programmable vertex and pixel shaders... To me, 15x more powerful isn't the average of the individual parts and that any one part has to be equal to or greater than 15x the previous part. But then maybe, that's just me... "powerful" is a bit of a subjective term.
BTW, the memory architecture is unified memory, not the typical "shared" memory that you find on MB chipsets. Both CPU and GPU can access the same exact memory locations. The CPU could twiddle a texture, then the GPU could directly use it for rendering. It won't have to pass over a bus first unlike a MB GPU that segregates a chuck of system ram to become it's own. It would have been easier, I think, to give the both the CPU and GPU their own 256Mb of RAM, but MS chose unified memory because it's more flexible for the developer and not the burden you seem to think it is.
I *was* just comparing straight Ghz to Ghz after all.
My bad, I've always been bored by the "watch a blank screen while a box counts to a million 15X faster than an older box" game.
The person I was replying to made some inane comment that it would take an 11ghz processor to be 15 times faster than an X-Box.
If what they're saying is "they can take old Xbox games (recompiled, obviously) and run then 15X faster on the new Xbox 360", they're lying. It just won't happen. The latency and bandwidth has not improved enough for that to be true, and the architecture is not easily exploitable for game software.
Remember, the architecture is the same on the X-Box 1. Shared memory. Only this time, there's more of it, the memory is much faster, the CPU can push data directly to the GPU (bypassing memory) and the GPU has embedded dram. Oh, and we're comparing a suped up G-Force 3 to something more powerful than an ATI x850.
Of course the Xbox 360 couldn't take Halo and run it at 900 fps, there's a ceiling, as you said. However, how about Halo with 15x more polygons? Xbox 1 theoretical poly throughput is 150 million. 15x that is 1.8 billion... quite a bit lower than the theoretical peak of 6 billion on Xbox360. Seeing as Halo hardly pushes 150 million polygons, I'm pretty sure Xbox 360 could handle 15x as many textured, lit and shaded polies as the Xbox 1. You could translate that into more enemies and bigger and more detailed environments. I bet it could do all that without even needing more than one core (leaving the others to improve physics, gameplay and/or AI) as all of that has more to do with the GPU than the CPU. So what's *your* definition of more powerful?
Of course, what they're saying is "the Xbox 360 has a peak MFLOPs 15X higher than the original Xbox", which is true, but utterly meaningless when it comes to games. The PS3 will have an utterly insane peak MFLOPs number (something like twice as high as the Xbox 360) but it certainly won't run games twice as fast.
Of course, you're right. There's more to a console than MFLOPs, but I'm not sure that the 15x more powerfull comes from that number anyway... if it did, I think the number would be bigger. Much bigger... They're saying the Xbox 360 does a TeraFLOP, right? I'd be hard pressed to believe the 733Mhz Celery does 66 MegaFLOPs. In fact, I'd be pretty suprised if it could do much more than 1 MegaFLOP... ZOMG! Xbox 360 is 1000x more powerful than the Xbox 1!!!!!111!!!!
Oops, sorry, got carried away...:-P
Anyway, not even comparing Mhz, MFLOPS or any other theoretical number, we've got Rare, running on alpha hardware (purportedly 1/3 the power of the final Xbox 360), creating a game that can handle 150 enemies with more polygons, more effects and a broader expanse of scenery that's all moving when they could barely scrape together 30 on the original xBox. http://xbox360.ign.com/articles/616/616768p1.html
I think all 3 new consoles are going to be a large leap in performance over their predessesors.
you ever look at a benchmark for a dual core CPU? They aren't even close to 100% faster.
Yes, I have. Have you? Depending on the benchmark, there were a few that offered around 80-90% improvements. http://techreport.com/reviews/2005q2/athlon64-x2/i ndex.x?pg=8 Specifically in 3d rendering. Compare the Athlon 4000 to the Athlon 4800 x2. I believe the 4000 even has a raw clock advantage here.
the XBOX 360 is going to have latency out the @ss due to the shared memory for graphics (ever used onboard PC graphics with shared memory, yeah it sucks
Ah, you *did* know that the X-box 1 has shared memory as well, didn't you? The X-box 360 has two advantages over the X-box, however. One, assuming they haven't changed the original announcement on the processor, the CPU can feed data directly to the GPU without going through memory and two, the GPU has 10 megs of EDRAM (embedded D-Ram).
BTW, you keep comparing things against the Revolution... I was just trying to point out that saying you would need an 11Ghz 3 core processor to beat a 733Mhz Celeron was, at best, an exaggeration.
True, we were comparing Ghz alone as the original poster did.
So you're saying that a 2 Ghz PPC970 isn't 2x more powerfull than a 1 Ghz Celeron? Even considering the speed of the memory coupled with the processor? Think carefully before you answer:-) The 733Mhz chip in the X-Box is a beefed up Celeron. Does it even have MMX-2? I know the X-Box cores aren't going to be PPC970, but I just wanted to point out the fallacy of your critisism.
Yah, sure, it's twice as fast at counting to a million. But it's not going to be twice as fast at running a generic program, not unless the memory latency halved and the throughput doubled. Anyone who's used a modern computer knows this.
Ah, I see you agree completely with what I said, then:-) I *was* just comparing straight Ghz to Ghz after all.
Nintendo just isn't giving you la-la land numbers. They never have. The fill rate for the N64 was quoted as 150K triangles. The one for the PS1 is quoted as 1.5M triangles.
The PS1 is not 10X more powerful than the N64. Sony quotes raw triangles, Nintendo quoted textured triangles.
Er... when did I say anything about Nintendo? Or Sony for that matter?
Anyway, if you look at the entire x-box system and compare it to the entire x-box 360 system, 10-15 times more overall power doesn't seem too far-fetched. If you go by quotes from the Rare about Kameo, it may be even more, depending on your metric. Rare has scenes running 3000 characters on screen in Kameo on Alpha hardware. Kameo was originally slated to be an X-Box game. At around 90% completion, they scrapped the code and re-wrote it from the ground up on the 360. The original X-Box game could only pull around 150 characters in that same scene, with lower poly characters and fewer effects.
Do I think comparing the raw specs is very meaningful? No. Do I think the difference between the 3 next gen consoles is going to be big? Not really. Especially in the first 2 years. Do I think the difference between this gen and next is going to be big? Yes. 10-15x bigger? Maybe.
That's comparing Ghz alone. Add in the more efficient cores, the dual thread execution and take out a bit for the inefficiencies in using multiple cores and 15x raw CPU power doesn't seem all that exaggerated.
Now, what the developers *do* with that extra power is something else all together. Me, I'm hoping for better A.I. beyond the usual "Memorize this pattern to beat the boss" that I see in > 90% of games today.
Re:For anyone who is sick and tired of MMORPGs...
on
Guild Wars Launches
·
· Score: 1
Four people in an instanced zone is hardly 'massive'.
There are definately zones with 6 people in a group and there may be some with 8 (haven't been everywhere, yet). But you're right, it's not really a 'massively' multiplayer online RPG. In fact, the Guild Wars website pretty much tells you the difference between it and an MMORPG. They refer to Guild Wars as a <market speak>Competitive Online Role Playing Game CORPG </market speak>.
In the first open betas, you could party with up to 4 people in the "starter" towns and 6 people in the later towns (never got above 12th level, so I don't know if there are towns with larger parties or not).
A couple of months ago, they added a whole "newbie" section of the game. In the first area, you can only party up to two players. Once you hit 3rd level you can move on to the 4 person per party areas.
Through all of this, the grandparent is correct, each party gets their own instance of the world. Only the towns are shared, and then they are broken up by servers (depending on the number of people trying to be in that particular town). Any other characters running about outside of town are NPCs or monsters.
This was several years ago, but when Adobe bought Aldus, I was offered an upgrade path from Photostyler to Photoshop. I *think* it was slightly higher than the Photoshop to Photoshop version upgrade, but only by around $20 or so.
They never did incorporate all of the cool features from Photostyler into Photoshop... and it wasn't until version 5 or 6 that I felt it was good enough to be a replacement for Photostyler. But then Photostyler was a Win31 program, so I couldn't keep using it forever:-P
Just to clarify the other poster's point, the Rio Karma is not a mass storage device, but the transfer utility is stored on the player, is written in Java and can be accessed through a web page through the ethernet port. So, it does work on Linux (assuming you have a JRE installed), but it isn't as nice as just mounting a Fat32 partition.
The Karma's a nice player, but it does have some drawbacks. The ethernet port is slow and not as compatible as it could be (it's mostly implemented in software on the player itself), the wheel wasn't as well engineered as it could have been, they have a few HD issues (in fact, mine just died a couple of weeks ago) and the warranty is pretty slim (90 days).
On the plus side is Flac, Ogg and gapless MP3 playback, an ethernet port and RCA jacks on the dock (included), 16 hour battery life (with MP3), the indescribably awesome Rio DJ, a five band grafic eq. with a few presets as well as 3 different user definable ones and a nice, clean audio out.
If you do buy one, this is one piece of electronics I'd recommend buying an extended warranty on.
1) The developments so far have all been browser based. No SKD necessary, just a copy of Wipeout Pure (and later, hopefully, Sony will release firmware with the browser built in).
2) Sony has twice released homebrew SDK's in the past. Yaroze! for the PS1 and the Linux development kit for the PS2. I would not be surprised if they did the same for the PSP (limited, of course, as the other two SDKs were).
In other words, the only problem I forsee Sony having with the "hacks" so far is the multigame "hack" (allowing multiple people to play a wireless game with only one copy of the game). The only thing I expect to happen with this is that developers will either embrace it (explicitly allow this to happen) or code their games so you can't do it.
If you've got a bright pixel (stuck on), then it's definately a LCD defect. Dark pixels on the PSP aren't necessarily LCD problems, though. If you tilt the unit side-to-side you might notice that the "dark pixel" moves across the front of the screen slightly. Apparently, there are defects in the clear plastic in front of the screen that makes some of the pixels appear dead. Add in the gratuitous amount of dust under the screen and it can really look like the LCD's are crap.
I unscrewed the front of the case from mine and blew some air under the cover (didn't take it completely off as that voids the warranty). Some of the sub-pixels I thought were dead turned out to be just dust. The other spots appeared to be in the clear window of the case itself.
I'm not saying this is the case for everyone, but it seems to be the problem with mine.
Actually, PSP games cost $40-$50. Still more expensive than DS games, but don't try to stretch the truth when the actual truth will do quite nicely.
Me, I'm just not interested in the DS lineup. A friend has one and there just aren't any games I'd like to play. The PSP, on the other hand, has three or four launch titles I'm interested in (all $40, btw). Will I bite? dunno... Guess I'll find out on Thursday:-)
If you read what the developers said in the forums, they received no notice of a Gold Master Candidate. Atari shipped without telling them, essentially, and plucked an older build to do it. It would be hard to believe that Atari tested it all that thoroughly as what they shipped was incomplete, buggy, and I'm not even sure you can complete the game without the first patch.
Don't shoot the messenger. I'm just parroting what the Devs were saying in their forums.
Gee, following that logic that means I can go buy a copy of Linux at Borders and utilize the source in my proprietary product with impunity.
I probably shouldn't respond to this, but... All software is protected by copyright. This is a limitation on distribution. I can buy Windows at the store and do just about anything but distribute copies of it legally. The same applies to Linux. Linux, however, comes with a license that states (among other things) that you are allowed to make copies of it if you agree to some terms (distribute source changes, etc.). This isn't an EULA that *RESTRICTS* your lawful rights, but a license that gives you *MORE* rights than you would normally have. If you don't agree to the license and distribute copies anyway, than you are in violation of copyright. Simple, eh?
According the the Troika staff, Atari published before they were finished and the build they used was not the build Troika was working on. Perhaps "published early" wasn't quite the right phrase, but the Troika staff (well, programming staff, anyway) was not expecting the product to ship when it did. Of course, I guess you could just say they are lying...
The first patch was basically the stuff they had already completed since the build that Atari published. After that, one employee, working at home, worked out some of the showstopper bugs and fans fixed a whole bunch of others.
Slashdot Mirror
In practice, perhaps. I didn't deny that. You *could* however, actually do that in a unified memory structure but not in a shared memory configuration. Is this an advantage? Perhaps, perhaps not. You want 400 megs of textures? You got it. No difference from running 100 megs of textures.
MS brought a console to market, with games, in a single year. They also didn't have the advantage of hardware relationships and none of the hardware components were designed in-house. It wasn't so much that they designed it poorly, they did the best they could in the short amount of time they had because they had to use components which were already researched and tested. Compare this to PS2 which was in the making for at least 3 years while Sony tinkered with the EE to get it up and running. They own practically everything, so it's trivial to manufacture it on any process they want, even combining dies if they need to.
I'd like to point out that's real-world measured latency, not absolute latency. RDRAM had a latency of around 11 clocks on first read, if I remember correctly. I'm not sure what DDR had at the time, but I'm sure it wasn't more than 4.
Remember, we're putting the CPU on the GPU memory bus and not the other way around. Sure, your solution is going to be faster. A ton faster? I don't think so. The reality of it though is that you'd really be choosing between this: shared memory bandwidth and 10M of EDRAM that will drop siginificantly in price or segragated memory busses that won't drop in price and no EDRAM. I'm guessing it's about equal once you factor 4xMSAA into the mix except for the fact the first solution will be cheaper in the long run.
Yes, that is correct, but they don't generally spec ram latency in ns, they spec it in clocks. Cas 3-4-4-8, isn't 3 ns. It's 3 clock cycles. I believe GDDR3 is 5-6-7-9 or something like that (who knows what's actually in the 360). So yeah, you're doubling the latency in some situations, but you're transferring data at a much higher rate to more than make up for it. RDRAM, btw, had a much higher latency than that when it first came out and DDR2 isn't all that great either, when compared to DDR1. So yeah, there's a penalty, but in real world applications, it's not going to be much more than 10-20% compared to equally clocked DDR1 (not that you could clock it that high).
Which is where great programmers will make the difference over good programmers. Take a look at GT4 and God of War for PS2. Do those games look great because the PS2 is a well balanced machine? No, they look good because the programmers know the hardware inside and out and can schedule data flow to it's maximum effect. I'd also say that the 360 is a better balanced machine than
According to you, the GeForce 6800 Ultra is only about 4x more powerfull than a Voodoo 5-5500 and only 25% faster than the onboard video of the 925 chipset because the 6800 Ultra can only run Quake 3 at 400fps VS the Voodoo 5's 100fps and the 925's 300fps? Actually, it's not even that because this is only 4x the framerate, not 4x the polygons, 4x the texture resolution, 4x the whatever else you decide to ignore as a factor. Oh and I've completely ignored the fact the 6800 Ultra is running at 1280x1024 and not 1024x768 because that's not a factor either.
BTW, "a whole of it's parts" would be having a fruit basket with 8 fruit because you have 4 apples and 4 oranges (and a basket). "A sum of it's parts" would be having 4 apples and 4 oranges and adding them to get 8 fruit. The real term you are looking for is "The median of it's parts," or "it's only as strong as it's weakest link."
Which, by the way, isn't found by looking at the bandwidth and seeing that the 360 only has 3.5 times as much. The two systems use bandwidth completely differently and we really won't know for a couple of years if that's really the weak point of the system or not. Especially since 4xMSAA uses 0% of the main system bandwidth and the 10MB of EDRAM is on a seperate bus (caching texture, z-buffers, frame buffers and fragment programs).
True, and we saw this with RAMBUS on the PC not long ago... of course, we're talking about 3.5 times increase over bus speeds for today's modern processors and the P4 didn't suffer *that* much from high latentcy RAMBUS memory in real world applications. Heck, dual core P4's will be in a worse situation, bandwidth-wise, than the 3 core 360 processor.
In the first year or two, I'm sure you'll be right. After that? I'm not so sure.
Maybe you have a differen't definition of segregated than I do. The CPU is in no way limited from writing/reading anywhere in system memory and neither is the GPU. How is that segregated?
GART sucked. Notice how both ATI and Nvidia are going back to this now that we have PCIe? Of course, main system bandwidth sucks (6.4GB/s), so it isn't all that usefull, yet.
But this also brings up the 10MB of EDRAM in the GPU you keep ignoring. Tech Report http://techreport.com/etc/2005q2/xbox360-gpu/index .x?pg=1 has an interesting article on the internals of the GPU. Not only is the 10MB used as a cache, but the circuitry surrounding the memory itself can do 4x MSAA for "free." In essence, the GPU isn't going to be going to main system RAM nearly as much as a typical PC Video card will. Not only that, there is no bandwidth required from the main system memory to do MSAA.
I was thinking dual-ported ram, as would be the straight-forward way of doing this. The GPU is also the memory controller, though, so it is a bit cheaper. BTW, you still need two busses, one from the CPU to the GPU and one from the GPU to the Memory, so it's not as huge a savings as you make it sound. You basically save the cost of needing two memory controllers instead of one.
er... the whole of its parts and the sum of its parts is, well, the same thing... In any case, I know wha
But, let's go back to your numbers. If the Xbox 360 could do 400 M/s and the xBox 1 could do 100 M/s (just rounding to make things easier), but could texture them the same, would that make the Xbox 360 4x more powerful? Now add in 4x as many textures... how do you measure that?
In my opinion, you don't have to do 15x the resolution to be 15x more powerful. As I see it, 720p is 3x the pixel count of 480p. Add 4x the polly count (have you seen the alpha games like Test Drive?) and 4x the texture quality, where does that leave you? And that's just basic graphics that doesn't even take into account what you can do with programmable vertex and pixel shaders... To me, 15x more powerful isn't the average of the individual parts and that any one part has to be equal to or greater than 15x the previous part. But then maybe, that's just me... "powerful" is a bit of a subjective term.BTW, the memory architecture is unified memory, not the typical "shared" memory that you find on MB chipsets. Both CPU and GPU can access the same exact memory locations. The CPU could twiddle a texture, then the GPU could directly use it for rendering. It won't have to pass over a bus first unlike a MB GPU that segregates a chuck of system ram to become it's own. It would have been easier, I think, to give the both the CPU and GPU their own 256Mb of RAM, but MS chose unified memory because it's more flexible for the developer and not the burden you seem to think it is.
Of course the Xbox 360 couldn't take Halo and run it at 900 fps, there's a ceiling, as you said. However, how about Halo with 15x more polygons? Xbox 1 theoretical poly throughput is 150 million. 15x that is 1.8 billion... quite a bit lower than the theoretical peak of 6 billion on Xbox360. Seeing as Halo hardly pushes 150 million polygons, I'm pretty sure Xbox 360 could handle 15x as many textured, lit and shaded polies as the Xbox 1. You could translate that into more enemies and bigger and more detailed environments. I bet it could do all that without even needing more than one core (leaving the others to improve physics, gameplay and/or AI) as all of that has more to do with the GPU than the CPU. So what's *your* definition of more powerful?
Of course, you're right. There's more to a console than MFLOPs, but I'm not sure that the 15x more powerfull comes from that number anyway... if it did, I think the number would be bigger. Much bigger... They're saying the Xbox 360 does a TeraFLOP, right? I'd be hard pressed to believe the 733Mhz Celery does 66 MegaFLOPs. In fact, I'd be pretty suprised if it could do much more than 1 MegaFLOP... ZOMG! Xbox 360 is 1000x more powerful than the Xbox 1!!!!!111!!!!Oops, sorry, got carried away... :-P
Anyway, not even comparing Mhz, MFLOPS or any other theoretical number, we've got Rare, running on alpha hardware (purportedly 1/3 the power of the final Xbox 360), creating a game that can handle 150 enemies with more polygons, more effects and a broader expanse of scenery that's all moving when they could barely scrape together 30 on the original xBox. http://xbox360.ign.com/articles/616/616768p1.html
I think all 3 new consoles are going to be a large leap in performance over their predessesors.
BTW, you keep comparing things against the Revolution... I was just trying to point out that saying you would need an 11Ghz 3 core processor to beat a 733Mhz Celeron was, at best, an exaggeration.
So you're saying that a 2 Ghz PPC970 isn't 2x more powerfull than a 1 Ghz Celeron? Even considering the speed of the memory coupled with the processor? Think carefully before you answer :-) The 733Mhz chip in the X-Box is a beefed up Celeron. Does it even have MMX-2? I know the X-Box cores aren't going to be PPC970, but I just wanted to point out the fallacy of your critisism.
Ah, I see you agree completely with what I said, thenAnyway, if you look at the entire x-box system and compare it to the entire x-box 360 system, 10-15 times more overall power doesn't seem too far-fetched. If you go by quotes from the Rare about Kameo, it may be even more, depending on your metric. Rare has scenes running 3000 characters on screen in Kameo on Alpha hardware. Kameo was originally slated to be an X-Box game. At around 90% completion, they scrapped the code and re-wrote it from the ground up on the 360. The original X-Box game could only pull around 150 characters in that same scene, with lower poly characters and fewer effects.
Do I think comparing the raw specs is very meaningful? No. Do I think the difference between the 3 next gen consoles is going to be big? Not really. Especially in the first 2 years. Do I think the difference between this gen and next is going to be big? Yes. 10-15x bigger? Maybe.
733Mhz * 15 = 11 Ghz (approximately)
3.2Ghz * 3 Cores = 9.8 Ghz
That's comparing Ghz alone. Add in the more efficient cores, the dual thread execution and take out a bit for the inefficiencies in using multiple cores and 15x raw CPU power doesn't seem all that exaggerated.
Now, what the developers *do* with that extra power is something else all together. Me, I'm hoping for better A.I. beyond the usual "Memorize this pattern to beat the boss" that I see in > 90% of games today.
A couple of months ago, they added a whole "newbie" section of the game. In the first area, you can only party up to two players. Once you hit 3rd level you can move on to the 4 person per party areas.
Through all of this, the grandparent is correct, each party gets their own instance of the world. Only the towns are shared, and then they are broken up by servers (depending on the number of people trying to be in that particular town). Any other characters running about outside of town are NPCs or monsters.
Oh, wait a minute...
They never did incorporate all of the cool features from Photostyler into Photoshop... and it wasn't until version 5 or 6 that I felt it was good enough to be a replacement for Photostyler. But then Photostyler was a Win31 program, so I couldn't keep using it forever :-P
The Karma's a nice player, but it does have some drawbacks. The ethernet port is slow and not as compatible as it could be (it's mostly implemented in software on the player itself), the wheel wasn't as well engineered as it could have been, they have a few HD issues (in fact, mine just died a couple of weeks ago) and the warranty is pretty slim (90 days).
On the plus side is Flac, Ogg and gapless MP3 playback, an ethernet port and RCA jacks on the dock (included), 16 hour battery life (with MP3), the indescribably awesome Rio DJ, a five band grafic eq. with a few presets as well as 3 different user definable ones and a nice, clean audio out.
If you do buy one, this is one piece of electronics I'd recommend buying an extended warranty on.
1) The developments so far have all been browser based. No SKD necessary, just a copy of Wipeout Pure (and later, hopefully, Sony will release firmware with the browser built in).
2) Sony has twice released homebrew SDK's in the past. Yaroze! for the PS1 and the Linux development kit for the PS2. I would not be surprised if they did the same for the PSP (limited, of course, as the other two SDKs were).
In other words, the only problem I forsee Sony having with the "hacks" so far is the multigame "hack" (allowing multiple people to play a wireless game with only one copy of the game). The only thing I expect to happen with this is that developers will either embrace it (explicitly allow this to happen) or code their games so you can't do it.
I unscrewed the front of the case from mine and blew some air under the cover (didn't take it completely off as that voids the warranty). Some of the sub-pixels I thought were dead turned out to be just dust. The other spots appeared to be in the clear window of the case itself.
I'm not saying this is the case for everyone, but it seems to be the problem with mine.
Me, I'm just not interested in the DS lineup. A friend has one and there just aren't any games I'd like to play. The PSP, on the other hand, has three or four launch titles I'm interested in (all $40, btw). Will I bite? dunno... Guess I'll find out on Thursday :-)
Don't shoot the messenger. I'm just parroting what the Devs were saying in their forums.
The first patch was basically the stuff they had already completed since the build that Atari published. After that, one employee, working at home, worked out some of the showstopper bugs and fans fixed a whole bunch of others.