Given that the SSDs are very nearly striped at the block level anyway, I can't imagine that RAID 0 is adding much more than flakiness.
I actually tried both single drive and RAID0 in my Vista configuration. Single drive took about 20 seconds to boot (once POST completes), RAID0 takes about 10 seconds to boot. So it's twice as fast based on a simple real world timings (VISTA boot speed).
For about $900, or the cost of the Fusion ioXtreme 80GB card, I bought two Intel 160GB SSD drives that I have in a RAID 0 configuration. It's very fast and 4X the capacity for the same price. Oh, and it's bootable.
No, that major reason is publisher effort, and the inconsistency of video driver support. And the first will never happen without the latter. Oh, and most games being written in DirectX.
Kudos to anyone still writing in OpenGL.
Most games are written on a Cross Platform Game Engine (like Unreal) nowdays. Underneath the hood is OpenGL or DirectX on a PC (or XBOX 360) but if you target PS3 or Wii, the game engine has to target a different rendering library.
As long as the terrorists are superstitious or believe they work, then they will actually work to some degree. For two reasons: 1) They will discourage people taking bombs through checkpoints with the devices. 2) A person transporting a bomb through a checkpoint is much more likely to act nervously or flighty in the presence of a "bomb detector" even if it doesn't work -- thus flagging them for further inspection.
I agree. I never said universal binaries were the problem -- rather, it's the lack of binary compatibility. It's the whole "source compatibility" / "binary incompatibility" that gives the win to Windoze over Linux in the gamine world.
Giving away free (gratis) access to some proprietary technology is nothing more than a complex marketing ploy to try to attract more commercial licensee in the long term, by gaining more fans and hackers in the short term. The basic idea is "let the Indie market play around with the engine, and if some group emerge with a new killer-app, they'll have to license our engine".
Whereas giving complete freedom to tinker with the GPL is the most community enabling. Granted, id Tech 5 is not in the GPL now. But on the other hand, the full freedom offered by the GPL has enabled heavy customisation such as the above and many other. And in the long term, are much more valuable for creativity.
Lets say someone makes a Killer-App (or more likely a kick-ass demo that will lead to further funding to complete the game) with either of these engines.
With Epic's Unreal Engine, they can make a game that can target PS3 and XBOX 360 markets which are significantly larger than the PC market. The can also make a closed source commercial game that targets the PC market. It's possible that Epic even intends to use this as an incubator to fund projects. Killer App --> Potential Profit.
With a GPL game engine, they have to open-source all of their code (hence not allowing them to target consoles). They can't even just license the several year old engine technology from ID if they want to go closed source (i.e. for PS3 / XBOX dev) because there have been other contributions. They basically have to "give away" the game for free since once you've released full source code for your game, there is pretty much no reason to buy it. The only chance for profit is by doing stuff like server based in-game content purchases (i.e. Second Life). Killer App --> ??? --> Profit -- if you can fill in the "???" part, you'll make me believe it's better to use the GPL engine. I guess the "???" part could be do all the work over in a closed-source engine and then ship but at that point, why not just start with a closed-source engine.
Now, if there was a high-end BSD-licensed game engine (as opposed to GPL), then this discussion would be moot and I'd be in agreement to use the BSD engine.
In the entire forked-up mess of the unix tree, there was only one thing that anybody & everybody cared about - source compatibilty. C99, POSIX, SuS v3, so many ways you could ensure that your code would compile everywhere, with whatever compiler was popular that week.
This guy worked in the closed-source world of video games where it's often not even legal to share your source code (due to middle-ware licensing and trade secrets) and even when it is legal, it's often not feasible for business or gameplay reasons (competitive coding advantage, preventing cheating hacks, disallowing "free content" mods, etc). It's exactly this reason that high-end cutting-edge games and other closed-source software will NEVER be viable on Linux unless there are major changes to the entire model of gaming development.
Security patches are not subject to the Genuine Advantage check.
If you turn on automatic updates, it will automatically install "Genuine Advantage" as well as security updates. Once you have "Genuine Advantage" on a machine that it thinks isn't "Genuine" -- it starts doing intrusive stuff -- From Wikipedia
Microsoft began distributing Windows Genuine Advantage Notifications as "critical update" KB905474 to Windows users. Users with pirated copies were exposed to alerts at startup, login, and during use of the Windows OS, stating that they do not have a genuine copy of Windows. Users with legitimate copies are not supposed to see the alerts (although some do anyway).
This also affected quite a few people who bought machines where they thought they had a legitimate copy of Windows but the one installed either wasn't legit or just had a common manufacturer key.
I believe the term for harvesting the power of the sun is solar energy.
And yes the Sun's energy is original from Fusion but under wildly different circumstances (crushing gravitational forces vs magnetic confinement).
I don't know any major problems with the Intel G1 SSD firmware but this is the second big issue with the G2 firmware. When the G2 drives first shipped, a bug in the firmware made it so if you changed the password, you could lose all your data.
I paid some big $$$ for two of the original first generation Intel 160GB drives which I have in a RAID 0 configuration. I'm a bit disappointed that Intel will not be offering a BIOS update to support TRIM commands for the G1 drives.
FWIW, Having Strippers at the company is *NOT* a good idea. The last game company I worked for had a stripper come in for the art directors birthday. I t was very awkward -- especially since she tried to get him to strip as well (which is something I did not need to see). Plus between married guys and nerds, no one really knew what we should be doing (I guess neither married guys nor nerds get sex).
Oh, and to top it all off, the one woman who was working there at the time (the receptionist) ended up suing the company for sexual harassment when she quit.
Not only did it have DMA, but cleverly the memory ran at 2x the speed of the CPU, with alternate cycles being given to the CPU and custom hardware.
This is only half true. The DMA did have priority access to memory on alternate cycles. However, the memory did not run 2X faster than the CPU.
The first 1MB of memory (CHIP MEM) was considerably slower (up to 50%) to access than higher addressed memory (FAST MEM) if you had a memory expansion due to the fact that CHIP MEM shared cycles with DMA and DMA took priority over the CPU.
Typically, when you've swapped out a process on the CPU, most of the L1 cache is going to be evicted by the next process which runs. A considerable amount of data will remain in the L2 cache but the L1 cache will be quite polluted by the intermediate processes that run between time slices. Therefore, L2 is the more important consideration.
The system they performed testing on has a CPU with a single shared L2 cache so processes moving between CPU cores are not necessarily slower than processes parked on a single core. Since the L2 is shared, there are no L2 cache misses introduced by swapping cores.
A CPU that would better show performance improvements on Windows 7 by higher thread affinity would be one with a split L2 cache like the Intel Kentsfield Core 2 Quad Q6600. The Q6600 would have a lot of L2 cache misses on Vista and XP that would be eliminated by the optimization in Windows 7.
The Amiga offered one helluva lot of bang for your bucks back in the day.
But come on. Even a mid-spec'ed Windows PC can handle genuine video editing, multi-track virtual recording studios with awesome soft synths and effects plug-ins, 24-bit colour to massive resolutions. All without having to work too hard in order to play nice with other apps and the OS itself.
Great in the day, but only interesting in a historical context.
(Ex-Amiga 500+ owner and
developer.)
The thing that interests me about the Amiga OS was its efficiency -- the ability to do all those things and multitask so long ago. An Amiga 500 had a 7MHz 68000 (which took 4-8 cycles minimum per instruction) that had around 1.5 MIPS CPU processing capability and no FPU and came with 512K of RAM and no CPU Cache.
The current mid-spec PC is 2GHz superscalar pipelined CPU with FPU with about 3,000 MIPS capability and comes with 2 MB of cache (plus 2GB of RAM). The CPU cache alone for a PC is 2-4X greater than the ENTIRE System RAM for an Amiga.
Current low-to-mid-range PC's are at least 2,000 times more powerful on paper than an Amiga 500 in raw MIPS and probably 200,000+ times higher in FPU FLOPS (if timing fmadds) -- with SSE and/or double precision that number could go up to nearly a MILLION times faster for FP performance.
All I'm saying here is we shouldn't just dismiss this as a bad thing until we've fully explored the legislative and societal implications a team of crime-fighting mutants with superpowers would have.
I disagree. Just because they recieved superpowers doesn't mean they'd use them to fight crime. We might end up with a bunch of new super-criminals.
Thanks for telling all of us that the best measure of hardware's performance ingame is... to benchmark it with a game.
The drivers do actually accelerate a fair number of popular games -- in other words, you get a faster frame rate. However, the acceleration is based on a database (list of exe names) that have been tested to work with the drivers. Intel is continuing to add to the list as they test new games and tweak the drivers but they haven't enabled it as a general feature because it does not work correctly on 100% of titles out there.
If a driver makes my game go faster, I don't consider that cheating on the driver as long as they are honest about what they are doing. Dynamic-Load balancing between the GPU and CPU is only going to become more common in the future with OpenCL (physics/GPGPU/etc) and when GPU's like Larrabee (which can run Intel binary code) become more available.
However, for the purpose of being able to test GPU alone, the driver should have an option to disable the dynamic load balancing feature.
That would also make Windows a 0x10000000bit wrapper around a 0x1000000bit implementation of a 0x100000bit extension for a 0x10000bit patch to an 0x1000bit operating system, originally coded for a 0x100bit microprocessor, written by a 0x10bit company, that can't stand 0x1bit of competition.
Putting numbers in hex (0x notation) doesn't make this a "programmer joke". Especially if you get your base wrong since you're obviously trying to use binary in which case the value 128 would be represented as 0b10000000 or 10000000b, not 0x10000000 (which is actually 256M or 268,435,456 in decimal).
Here's a programmer joke (which sounds better spoken than read): To be or not to be.... equals 0xff.
Because of course the infrastructure to serve 3gb of data to each customer doesn't cost anything?
They already paid MS for Vista + the "FREE" Win7 upgrade. And the main reason people haven't stopped buying Vista PC's in the past couple months rather than waiting until Windows 7 came out is that they were promised FREE upgrades to Windows 7. I'm sorry but when I'm promised something for FREE along with a $1,000 purchase, I get a bit irritated when they charge me for the "FREE" item I was promised even if it's only $17.
Slashdot Mirror
Given that the SSDs are very nearly striped at the block level anyway, I can't imagine that RAID 0 is adding much more than flakiness.
I actually tried both single drive and RAID0 in my Vista configuration. Single drive took about 20 seconds to boot (once POST completes), RAID0 takes about 10 seconds to boot. So it's twice as fast based on a simple real world timings (VISTA boot speed).
For about $900, or the cost of the Fusion ioXtreme 80GB card, I bought two Intel 160GB SSD drives that I have in a RAID 0 configuration. It's very fast and 4X the capacity for the same price. Oh, and it's bootable.
No, that major reason is publisher effort, and the inconsistency of video driver support. And the first will never happen without the latter. Oh, and most games being written in DirectX. Kudos to anyone still writing in OpenGL.
Most games are written on a Cross Platform Game Engine (like Unreal) nowdays. Underneath the hood is OpenGL or DirectX on a PC (or XBOX 360) but if you target PS3 or Wii, the game engine has to target a different rendering library.
As long as the terrorists are superstitious or believe they work, then they will actually work to some degree. For two reasons: 1) They will discourage people taking bombs through checkpoints with the devices. 2) A person transporting a bomb through a checkpoint is much more likely to act nervously or flighty in the presence of a "bomb detector" even if it doesn't work -- thus flagging them for further inspection.
I agree. I never said universal binaries were the problem -- rather, it's the lack of binary compatibility. It's the whole "source compatibility" / "binary incompatibility" that gives the win to Windoze over Linux in the gamine world.
More seriously :
Giving away free (gratis) access to some proprietary technology is nothing more than a complex marketing ploy to try to attract more commercial licensee in the long term, by gaining more fans and hackers in the short term. The basic idea is "let the Indie market play around with the engine, and if some group emerge with a new killer-app, they'll have to license our engine".
Whereas giving complete freedom to tinker with the GPL is the most community enabling. Granted, id Tech 5 is not in the GPL now. But on the other hand, the full freedom offered by the GPL has enabled heavy customisation such as the above and many other. And in the long term, are much more valuable for creativity.
Lets say someone makes a Killer-App (or more likely a kick-ass demo that will lead to further funding to complete the game) with either of these engines.
With Epic's Unreal Engine, they can make a game that can target PS3 and XBOX 360 markets which are significantly larger than the PC market. The can also make a closed source commercial game that targets the PC market. It's possible that Epic even intends to use this as an incubator to fund projects. Killer App --> Potential Profit.
With a GPL game engine, they have to open-source all of their code (hence not allowing them to target consoles). They can't even just license the several year old engine technology from ID if they want to go closed source (i.e. for PS3 / XBOX dev) because there have been other contributions. They basically have to "give away" the game for free since once you've released full source code for your game, there is pretty much no reason to buy it. The only chance for profit is by doing stuff like server based in-game content purchases (i.e. Second Life). Killer App --> ??? --> Profit -- if you can fill in the "???" part, you'll make me believe it's better to use the GPL engine. I guess the "???" part could be do all the work over in a closed-source engine and then ship but at that point, why not just start with a closed-source engine.
Now, if there was a high-end BSD-licensed game engine (as opposed to GPL), then this discussion would be moot and I'd be in agreement to use the BSD engine.
In the entire forked-up mess of the unix tree, there was only one thing that anybody & everybody cared about - source compatibilty. C99, POSIX, SuS v3, so many ways you could ensure that your code would compile everywhere, with whatever compiler was popular that week.
This guy worked in the closed-source world of video games where it's often not even legal to share your source code (due to middle-ware licensing and trade secrets) and even when it is legal, it's often not feasible for business or gameplay reasons (competitive coding advantage, preventing cheating hacks, disallowing "free content" mods, etc). It's exactly this reason that high-end cutting-edge games and other closed-source software will NEVER be viable on Linux unless there are major changes to the entire model of gaming development.
Security patches are not subject to the Genuine Advantage check.
If you turn on automatic updates, it will automatically install "Genuine Advantage" as well as security updates. Once you have "Genuine Advantage" on a machine that it thinks isn't "Genuine" -- it starts doing intrusive stuff -- From Wikipedia
Microsoft began distributing Windows Genuine Advantage Notifications as "critical update" KB905474 to Windows users. Users with pirated copies were exposed to alerts at startup, login, and during use of the Windows OS, stating that they do not have a genuine copy of Windows. Users with legitimate copies are not supposed to see the alerts (although some do anyway).
This also affected quite a few people who bought machines where they thought they had a legitimate copy of Windows but the one installed either wasn't legit or just had a common manufacturer key.
I believe the term for harvesting the power of the sun is solar energy. And yes the Sun's energy is original from Fusion but under wildly different circumstances (crushing gravitational forces vs magnetic confinement).
I don't know any major problems with the Intel G1 SSD firmware but this is the second big issue with the G2 firmware. When the G2 drives first shipped, a bug in the firmware made it so if you changed the password, you could lose all your data.
I paid some big $$$ for two of the original first generation Intel 160GB drives which I have in a RAID 0 configuration. I'm a bit disappointed that Intel will not be offering a BIOS update to support TRIM commands for the G1 drives.
USB connected drives are painfully slow compared to SATA (or eSATA) connected drives. SATA HD's are quite a bit slower than a good SATA SSD.
However, I'm not sure how your post is on topic since your slow USB HD experience has very little to do with the longevity of SSD's ?!?
Can you at least tell us which 3 of your 8 drives failed ? Perhaps there is some similarity in controller or Flash memory used?
FWIW, I have 2 of the Intel Drives and 1 OCZ drive and I haven't seen any problems.
FWIW, Having Strippers at the company is *NOT* a good idea. The last game company I worked for had a stripper come in for the art directors birthday. I t was very awkward -- especially since she tried to get him to strip as well (which is something I did not need to see). Plus between married guys and nerds, no one really knew what we should be doing (I guess neither married guys nor nerds get sex).
Oh, and to top it all off, the one woman who was working there at the time (the receptionist) ended up suing the company for sexual harassment when she quit.
Interesting tidbit: Alinea provided a "legal" reason to allow herbal vaporizers to be imported into the US.
Not only did it have DMA, but cleverly the memory ran at 2x the speed of the CPU, with alternate cycles being given to the CPU and custom hardware.
This is only half true. The DMA did have priority access to memory on alternate cycles. However, the memory did not run 2X faster than the CPU.
The first 1MB of memory (CHIP MEM) was considerably slower (up to 50%) to access than higher addressed memory (FAST MEM) if you had a memory expansion due to the fact that CHIP MEM shared cycles with DMA and DMA took priority over the CPU.
Typically, when you've swapped out a process on the CPU, most of the L1 cache is going to be evicted by the next process which runs. A considerable amount of data will remain in the L2 cache but the L1 cache will be quite polluted by the intermediate processes that run between time slices. Therefore, L2 is the more important consideration.
The system they performed testing on has a CPU with a single shared L2 cache so processes moving between CPU cores are not necessarily slower than processes parked on a single core. Since the L2 is shared, there are no L2 cache misses introduced by swapping cores.
A CPU that would better show performance improvements on Windows 7 by higher thread affinity would be one with a split L2 cache like the Intel Kentsfield Core 2 Quad Q6600. The Q6600 would have a lot of L2 cache misses on Vista and XP that would be eliminated by the optimization in Windows 7.
The truth is we are all individuals.
I'm not.
You're unique, just like everyone else.
The Amiga offered one helluva lot of bang for your bucks back in the day.
But come on. Even a mid-spec'ed Windows PC can handle genuine video editing, multi-track virtual recording studios with awesome soft synths and effects plug-ins, 24-bit colour to massive resolutions. All without having to work too hard in order to play nice with other apps and the OS itself.
Great in the day, but only interesting in a historical context.
(Ex-Amiga 500+ owner and developer.)
The thing that interests me about the Amiga OS was its efficiency -- the ability to do all those things and multitask so long ago. An Amiga 500 had a 7MHz 68000 (which took 4-8 cycles minimum per instruction) that had around 1.5 MIPS CPU processing capability and no FPU and came with 512K of RAM and no CPU Cache.
The current mid-spec PC is 2GHz superscalar pipelined CPU with FPU with about 3,000 MIPS capability and comes with 2 MB of cache (plus 2GB of RAM). The CPU cache alone for a PC is 2-4X greater than the ENTIRE System RAM for an Amiga.
Current low-to-mid-range PC's are at least 2,000 times more powerful on paper than an Amiga 500 in raw MIPS and probably 200,000+ times higher in FPU FLOPS (if timing fmadds) -- with SSE and/or double precision that number could go up to nearly a MILLION times faster for FP performance.
All I'm saying here is we shouldn't just dismiss this as a bad thing until we've fully explored the legislative and societal implications a team of crime-fighting mutants with superpowers would have.
I disagree. Just because they recieved superpowers doesn't mean they'd use them to fight crime. We might end up with a bunch of new super-criminals.
Thanks for telling all of us that the best measure of hardware's performance ingame is... to benchmark it with a game.
The drivers do actually accelerate a fair number of popular games -- in other words, you get a faster frame rate. However, the acceleration is based on a database (list of exe names) that have been tested to work with the drivers. Intel is continuing to add to the list as they test new games and tweak the drivers but they haven't enabled it as a general feature because it does not work correctly on 100% of titles out there.
If a driver makes my game go faster, I don't consider that cheating on the driver as long as they are honest about what they are doing. Dynamic-Load balancing between the GPU and CPU is only going to become more common in the future with OpenCL (physics/GPGPU/etc) and when GPU's like Larrabee (which can run Intel binary code) become more available.
However, for the purpose of being able to test GPU alone, the driver should have an option to disable the dynamic load balancing feature.
Sorry, that should be exactly 8-bits... not at least 8-bits. Now I feel like I'm debugging my jokes and that's not funny at all!
For the purposes of this joke, the hex data type is unsigned and at least 8-bits.... jeesh... this is why programmers don't tell jokes!
That would also make Windows a 0x10000000bit wrapper around a 0x1000000bit implementation of a 0x100000bit extension for a 0x10000bit patch to an 0x1000bit operating system, originally coded for a 0x100bit microprocessor, written by a 0x10bit company, that can't stand 0x1bit of competition.
Putting numbers in hex (0x notation) doesn't make this a "programmer joke". Especially if you get your base wrong since you're obviously trying to use binary in which case the value 128 would be represented as 0b10000000 or 10000000b, not 0x10000000 (which is actually 256M or 268,435,456 in decimal).
.... equals 0xff.
Here's a programmer joke (which sounds better spoken than read): To be or not to be
Because of course the infrastructure to serve 3gb of data to each customer doesn't cost anything?
They already paid MS for Vista + the "FREE" Win7 upgrade. And the main reason people haven't stopped buying Vista PC's in the past couple months rather than waiting until Windows 7 came out is that they were promised FREE upgrades to Windows 7. I'm sorry but when I'm promised something for FREE along with a $1,000 purchase, I get a bit irritated when they charge me for the "FREE" item I was promised even if it's only $17.