The thing would be slow as molasses if you tried to run PPC code that wasn't optimized for this design change.
I'm not arguing that is false. I am arguing that your original premise -- that the unit is incapable of running general purpose logic efficiently -- is false.
I am of course ignoring the "it could run OSX" arguement, because that is never going to happen. But it is hardly a crippled processor incapable of performing every day tasks.
You aren't going to be dropping a random linux kernel on the box; you're going to recompile for the hardware you're running on. It hardly has to be insanely specialized, nor is it limited to some niche number crunching role. You just have the feed the processor well organized code.
The only thing I've found on google are forum posts from a guy who heard from a guy who knows a guy... The "information" is false. Made up. Bullshit. Anyone who has an iota of sense can figure that out.
Launch titles use all three cores; it takes no effort to find dev interviews talking about which core they have doing what.
The hypervisor isn't something that "takes" a core to run -- it is hardware support for virtualization.
So, as I said before, unless you can state an authorative source, I call bullshit.
Anyway, there are reports that only one core is availble to intitial game developers, and one of the cores is strictly for M$ bullshit content protection TC such as the hypervisor, etc.
An inability to execute instructions out of order does not prevent PPC code from running, nor is it crippling. Additionally, having an instruction window is rather pointless when instructions can't be executed out of order.
You don't need these things with properly compiled code. But that's the trick; you can't take poorly compiled code and expect it to be performant.
Traditionally general purpose CPU's need the capability to execute out of order because they're running code that wasn't written for their current design -- they have to run code designed to run on x previous iterations of the processor. And traditionally, CPU manufacturers like to tout how their new processor runs x benchmark 50 times faster -- and they can't do that without out of order execution.
You don't accept them. You utilize the warranty and get it fixed.
It isn't reasonable to expect every unit to roll off the line and make it into your home flawless. Nothing would ever be make it out of the factory if that was the requirement.
It is reasonable to expect a company to stand by their product if you get one with a defect. By all accounts, that is happening; replacements arriving in 5 days, with everything handled via overnight shipping.
Justifying what I'm seeing here as "slashdot loves to bash Microsoft" IS acceptable, because the failure rates ARE less than the industry norm AND Microsoft's warranty service for problem units has been exceptional.
Vista will never end up running on a PPC. There isn't an established open "platform" for it, and I don't see Microsoft jumping into the OEM business anytime soon (poor profit margins)...
A 5% failure rate for an "appliance" type consumer electronics device is generally considered acceptable in the industry.
The kind of "stress test" you propose would be unprecidented in the history of consumer electronics. It would also be the easiest way to make sure your production line is as inefficient as possible (think about the scale of what it is you are proposing). Finally, it still won't catch problems that are the result of damage occuring during transit.
In a production line you depend on your suppliers giving you parts that work to your specifications, and you make sure that when you put them all together the unit passes a set of diagnostics. Any problem that gets past that process is handled by the product warranty.
Who wouldn't? Wouldn't you like a 100mb cache? Hell, wouldn't you like all of main ram to actually be as fast as an L2 cache?
This is a game console, not a multi-tasking general purpose PC. This is reflected in both the power a programer has over the hardware, and it is reflected in the fact it depends on the programmer to make up for cost saving measures implemented in the hardware.
You're thinking about thread scheduling the way a general purpose PC would schedule threads. It doesn't work that way in a game system. When talking about threads on a game console, we're talking about hardware threads. The programmer decides what core runs what logic, when it runs that logic; there aren't any context switches, and threads don't "change" cores when running.
The cache design on this chip is actually quite cool. If you're only going to use a piece of data once (or infrequently), you can instruct the processor to not store the data in the L2 cache (leaving data that you're using more freqently in the cache). Additionally, the L1 and L2 cache can hold separate data, and a load into the L1 cache doesn't require a storing that data into the L2 cache. When writing data out to main memory, a program can instruct the processor not to store it in the cache. Software can also reserve chunks of cache for whatever purpose it wants. Finally, a direct link between the CPU and the GPU can be established where the GPU reads data directly off of the L2 cache as the CPU generates it.
These abilitys make it possible to utilize L2 cache better than you would with any general purpose PC. However, it requires more thinking and more effort to do so. That's the price/performance tradeoff you get.
Yes, it does seem stupid if you don't know how digital sensors work.
Without the "SLR mechanism", a lot of technical compromises have to be made. The biggest thing you'd be able to relate to is probably response time -- it takes non-trivial amounts of time to clear the sensor and switch the sensor into picture taking mode.
If you've ever wondered why every single point and shoot camera has a bit of "lag" between hitting the shutter button and the camera actually taking a picture, this is why. (on some point and shoots, the lag time is greatly reduced if you disable the live preview)
"Reveal codes" was necessary because WP was so buggy/unusable that you NEEDED to look at the tags to get a document formatted correctly.
The reason Word doesn't have a "reveal codes" option is that Word does not store formatting information inline with the document in a tagged format. There are no codes to view; everything is stored in objects. If you analyze how options are presented to you and how they effect the document, it shouldn't take you too long to determine how those objects are organized.
$60 is the price that they think they can sell at and make the most profit. Not all next gen games are $60. First party titles for the 360 are $50, and xbox live titles (which are admitidly far simpler) range from $5 to $20.
We'll see if that pricepoint sticks; ultimately, if people buy it they'll continue to charge that kind of money for it.
While processor time is something to be concerned about, it isn't the most limited resource in the machine; the biggest thing a game programmer is fighting is use of bandwidth. It is actually faster to decompress on the fly than it is to spend all your cpu time waiting on memory reads while saturating the bus.
Most game data is compressed in some form, and a lot of time is spent dealing with compression. It reduces load times and allows you to pack more data in ram -- the processor is fast enough that it is more efficient to deal with most data in a compressed form than it would be to constantly hit the disc for it in an uncompressed form. Hell, it is preferable to actually calculate data on the fly as needed instead of reading a precalculated table in ram!
Sure, you can always measure it. But personally, do you really care if the efficiency of a fuel cell 'battery' for a laptop is 50% worse than a Li battery if you get runtimes 8 times longer than the Li battery? For the small amounts of power we're talking about here, efficiency isn't terribly relevent.
It is far, FAR more relevent when we're talking about using fuel cells in things like cars. That is still a long way off, unfortunately.
Not only that, but the pc version is shipping on only one disc.
All of the "oh no, the sky is falling in because HD games need higher res [foo]" talk is just rediculous. PC games have been running at > HD resolutions for years, and they still manage to ship on one or two CDs.
This is a replacement powersource for your laptop or other portable device. The purpose of these things is to eliminate the hours it takes to recharge a device. Efficiency, while a factor, is not a primary factor.
There are plenty of others stating they left it in place
Unfortunately for them, I don't believe most of them. They all describe the behavior you'd see and hear when rotating the unit, except they make a big fuss about how they didn't move it. If it were a defect in the drive, it would happen consistently to every game they played. These are the same people who drop something they buy on the floor and take it back claiming it came out of the box broken.
Do some things come out of the box broken? Yeah. Do most things come out of the box broken? Nope.
These drives aren't custom hardware. They're standard sata dvd drives like you'd install in a computer. When was the last time you heard of a dvd drive eating discs? That's what I thought.
Some stores clerks are reporting rather high return rates on 360 discs because of circular gouges in the surface
Are these the same clerks who claimed that Microsoft was recalling the Xbox 360? Are these the same clerks stating that no new units will be shipped from Microsoft until next March? Do I need to go on about bullshit made up by clerks? "Retail clerks" as a group are not reliable sources for information. "Retail clerk Bob" might be, but he'll have to prove himself reliable before I'll believe him.
I can't believe how many ways they seemed to have scr**ed the pooch with this launch.
Lets see, so far they've sold roughly 700-800 thousand units in about a week. They've got fanboys making shit up in forums. They've got problems with people who can't plug in cables all the way. They've got problems with idiots who don't use common sense and don't own up to it. They've got problems with idiots blocking exhaust vents with power supplies. They've got manufacturing defects with about 3% of units.
I don't know about you, but I don't see anything really out of line with what you'd expect for any product launch.
Ah, I see. If I call an idiot and idiot I/must/ work for the 360 team. Do you work for the PS3 team? Your perspective seems to be skewed; a proper attitude would be "yeah, the rotational inertia of something moving at seven thousand fucking RPM isn't something a reasonable person would expect a DVD drive to handle."
Common sense tells me this should withstand a reasonable amount of abuse.
Common sense tells me that my idea of a reasonable amount of abuse differs from yours. Common sense tells me that it is probably a bad idea to fuck with something while it is moving.
I bet you get pissed off that VCRs break if you put a pb&j sandwich in the slot too.
Slashdot Mirror
The thing would be slow as molasses if you tried to run PPC code that wasn't optimized for this design change.
I'm not arguing that is false. I am arguing that your original premise -- that the unit is incapable of running general purpose logic efficiently -- is false.
I am of course ignoring the "it could run OSX" arguement, because that is never going to happen. But it is hardly a crippled processor incapable of performing every day tasks.
You aren't going to be dropping a random linux kernel on the box; you're going to recompile for the hardware you're running on. It hardly has to be insanely specialized, nor is it limited to some niche number crunching role. You just have the feed the processor well organized code.
The only thing I've found on google are forum posts from a guy who heard from a guy who knows a guy ... The "information" is false. Made up. Bullshit. Anyone who has an iota of sense can figure that out.
Launch titles use all three cores; it takes no effort to find dev interviews talking about which core they have doing what.
The hypervisor isn't something that "takes" a core to run -- it is hardware support for virtualization.
So, as I said before, unless you can state an authorative source, I call bullshit.
Turnaround time is reported to be 5 days, including shipping time.
Anyway, there are reports that only one core is availble to intitial game developers, and one of the cores is strictly for M$ bullshit content protection TC such as the hypervisor, etc.
Authoritative sources please. I'm calling bullshit.
An inability to execute instructions out of order does not prevent PPC code from running, nor is it crippling. Additionally, having an instruction window is rather pointless when instructions can't be executed out of order.
You don't need these things with properly compiled code. But that's the trick; you can't take poorly compiled code and expect it to be performant.
Traditionally general purpose CPU's need the capability to execute out of order because they're running code that wasn't written for their current design -- they have to run code designed to run on x previous iterations of the processor. And traditionally, CPU manufacturers like to tout how their new processor runs x benchmark 50 times faster -- and they can't do that without out of order execution.
You don't accept them. You utilize the warranty and get it fixed.
It isn't reasonable to expect every unit to roll off the line and make it into your home flawless. Nothing would ever be make it out of the factory if that was the requirement.
It is reasonable to expect a company to stand by their product if you get one with a defect. By all accounts, that is happening; replacements arriving in 5 days, with everything handled via overnight shipping.
Justifying what I'm seeing here as "slashdot loves to bash Microsoft" IS acceptable, because the failure rates ARE less than the industry norm AND Microsoft's warranty service for problem units has been exceptional.
Vista will never end up running on a PPC. There isn't an established open "platform" for it, and I don't see Microsoft jumping into the OEM business anytime soon (poor profit margins) ...
A 5% failure rate for an "appliance" type consumer electronics device is generally considered acceptable in the industry.
The kind of "stress test" you propose would be unprecidented in the history of consumer electronics. It would also be the easiest way to make sure your production line is as inefficient as possible (think about the scale of what it is you are proposing). Finally, it still won't catch problems that are the result of damage occuring during transit.
In a production line you depend on your suppliers giving you parts that work to your specifications, and you make sure that when you put them all together the unit passes a set of diagnostics. Any problem that gets past that process is handled by the product warranty.
Who wouldn't? Wouldn't you like a 100mb cache? Hell, wouldn't you like all of main ram to actually be as fast as an L2 cache?
This is a game console, not a multi-tasking general purpose PC. This is reflected in both the power a programer has over the hardware, and it is reflected in the fact it depends on the programmer to make up for cost saving measures implemented in the hardware.
You're thinking about thread scheduling the way a general purpose PC would schedule threads. It doesn't work that way in a game system. When talking about threads on a game console, we're talking about hardware threads. The programmer decides what core runs what logic, when it runs that logic; there aren't any context switches, and threads don't "change" cores when running.
The cache design on this chip is actually quite cool. If you're only going to use a piece of data once (or infrequently), you can instruct the processor to not store the data in the L2 cache (leaving data that you're using more freqently in the cache). Additionally, the L1 and L2 cache can hold separate data, and a load into the L1 cache doesn't require a storing that data into the L2 cache. When writing data out to main memory, a program can instruct the processor not to store it in the cache. Software can also reserve chunks of cache for whatever purpose it wants. Finally, a direct link between the CPU and the GPU can be established where the GPU reads data directly off of the L2 cache as the CPU generates it.
These abilitys make it possible to utilize L2 cache better than you would with any general purpose PC. However, it requires more thinking and more effort to do so. That's the price/performance tradeoff you get.
Yes, it does seem stupid if you don't know how digital sensors work.
Without the "SLR mechanism", a lot of technical compromises have to be made. The biggest thing you'd be able to relate to is probably response time -- it takes non-trivial amounts of time to clear the sensor and switch the sensor into picture taking mode.
If you've ever wondered why every single point and shoot camera has a bit of "lag" between hitting the shutter button and the camera actually taking a picture, this is why. (on some point and shoots, the lag time is greatly reduced if you disable the live preview)
"Reveal codes" was necessary because WP was so buggy/unusable that you NEEDED to look at the tags to get a document formatted correctly.
The reason Word doesn't have a "reveal codes" option is that Word does not store formatting information inline with the document in a tagged format. There are no codes to view; everything is stored in objects. If you analyze how options are presented to you and how they effect the document, it shouldn't take you too long to determine how those objects are organized.
What is this, like appology number 12? Dang, how many times are they going to appologize for this?
$60 is the price that they think they can sell at and make the most profit. Not all next gen games are $60. First party titles for the 360 are $50, and xbox live titles (which are admitidly far simpler) range from $5 to $20.
We'll see if that pricepoint sticks; ultimately, if people buy it they'll continue to charge that kind of money for it.
While processor time is something to be concerned about, it isn't the most limited resource in the machine; the biggest thing a game programmer is fighting is use of bandwidth. It is actually faster to decompress on the fly than it is to spend all your cpu time waiting on memory reads while saturating the bus.
You'd be surprised. Read this: http://www.xbox.com/NR/rdonlyres/3FCB65F9-E9E5-45D B-B7F9-59A5EE265B8F/0/Xbox360Preparation.doc
Most game data is compressed in some form, and a lot of time is spent dealing with compression. It reduces load times and allows you to pack more data in ram -- the processor is fast enough that it is more efficient to deal with most data in a compressed form than it would be to constantly hit the disc for it in an uncompressed form. Hell, it is preferable to actually calculate data on the fly as needed instead of reading a precalculated table in ram!
Sure, you can always measure it. But personally, do you really care if the efficiency of a fuel cell 'battery' for a laptop is 50% worse than a Li battery if you get runtimes 8 times longer than the Li battery? For the small amounts of power we're talking about here, efficiency isn't terribly relevent.
It is far, FAR more relevent when we're talking about using fuel cells in things like cars. That is still a long way off, unfortunately.
If you're encoding with mpeg2 yeah. Not if you're encoding with some of the more modern video codecs.
Regardless of that point, I personally doubt I'll lose much sleep if the next rpg I get doesn't play like a choose your own adventure book.
Not only that, but the pc version is shipping on only one disc.
All of the "oh no, the sky is falling in because HD games need higher res [foo]" talk is just rediculous. PC games have been running at > HD resolutions for years, and they still manage to ship on one or two CDs.
Actually, there already is an option to disable it. The article is an entire knee-jerk.
Efficiency doesn't matter.
This is a replacement powersource for your laptop or other portable device. The purpose of these things is to eliminate the hours it takes to recharge a device. Efficiency, while a factor, is not a primary factor.
There are plenty of others stating they left it in place
Unfortunately for them, I don't believe most of them. They all describe the behavior you'd see and hear when rotating the unit, except they make a big fuss about how they didn't move it. If it were a defect in the drive, it would happen consistently to every game they played. These are the same people who drop something they buy on the floor and take it back claiming it came out of the box broken.
Do some things come out of the box broken? Yeah. Do most things come out of the box broken? Nope.
These drives aren't custom hardware. They're standard sata dvd drives like you'd install in a computer. When was the last time you heard of a dvd drive eating discs? That's what I thought.
Some stores clerks are reporting rather high return rates on 360 discs because of circular gouges in the surface
Are these the same clerks who claimed that Microsoft was recalling the Xbox 360? Are these the same clerks stating that no new units will be shipped from Microsoft until next March? Do I need to go on about bullshit made up by clerks? "Retail clerks" as a group are not reliable sources for information. "Retail clerk Bob" might be, but he'll have to prove himself reliable before I'll believe him.
I can't believe how many ways they seemed to have scr**ed the pooch with this launch.
Lets see, so far they've sold roughly 700-800 thousand units in about a week. They've got fanboys making shit up in forums. They've got problems with people who can't plug in cables all the way. They've got problems with idiots who don't use common sense and don't own up to it. They've got problems with idiots blocking exhaust vents with power supplies. They've got manufacturing defects with about 3% of units.
I don't know about you, but I don't see anything really out of line with what you'd expect for any product launch.
Rriiighttt ... that's why they've spent all that time litigating it trying to get the patent thrown out.
Ah, I see. If I call an idiot and idiot I /must/ work for the 360 team. Do you work for the PS3 team? Your perspective seems to be skewed; a proper attitude would be "yeah, the rotational inertia of something moving at seven thousand fucking RPM isn't something a reasonable person would expect a DVD drive to handle."
I loved that console. Too bad the gits running Atari at the time were nimrods.
Common sense tells me this should withstand a reasonable amount of abuse.
Common sense tells me that my idea of a reasonable amount of abuse differs from yours. Common sense tells me that it is probably a bad idea to fuck with something while it is moving.
I bet you get pissed off that VCRs break if you put a pb&j sandwich in the slot too.