I have to agree, Via Mini-ITX are a lackluster choice for desktop use.
The problem is the anemic processor bus, which makes any rich content seem sluggish. Thanks to the Nemiah core upgrade, the processor is perfectly capable, but any modern GUI interface (XP, GNOME, KDE) will feel slow. Overall performance under load also suffers, as it becomes bus-bound.
An upgrade to a 200MHz revamp of the GTL bus and a slightly larger L2 cache would make this chip competitve without sacrificing cost or performance. Now especially that Via is offering a combined C3 core and northbridge on a single package, there's no reason the bus couldn't be pumped up (similar to what Freescale did with their integrated 667MHz MPX bus on the dual core MPC8641D).
Unfortunately for us, the Via Epia is already overpowered for it's real market: kiosk, POS, and other low-power specialty applications benefiting from x86 compatibility. Via is happy to release variants with LVDS panel controllers and compactflash slots, but there is no market need for improving processor bandwidth.
So, get used to that poor desktop performance. Luckily, AMD and Intel are starting to intrude on Via's low-power computer enthusiast market with the Athlon 64 and the Pentium M.
Basically, the article is simply parroting what the Intel marketing department has been told by their engineers: that leakage current will be reduced by allmost 1000 times. The Intel marketing department is simply palming it off as some newfangled "ultra low power" process, when it's been on the roadmap the whole time.
That is, if the reality about high-K comes anywhere near Intel's claims. I recall quite clearly Intel skipping out on SOI, and claiming quite calmly to the world that strained silicon would solve all their problems in the move to 90nm. That didn't happen. Ultra-aggressive dynamic power-saving modes and the Pentium M are the only things that saved Intel's 90nm process.
Will be impressive if they actually get results in the same ballpark as their estimates, though.
I'd like to add that the Turion ML processor used in that test is part of the higher-power usage cores with a TDP of 35w.
The Turion MT processor (25w TDP) is available all the way up to 2.2GHz (since the end of August), and is QUITE competitve in terms of power usage with the Pentium M, even at full load.
They're evenly-matched in the innovation arena. If you folks really want to know why Apple chose Intel over AMD, I've already run the numbers. Though AMD is quire innovative and competitive, it would be risky for Apple to depend solely on AMD.
And? There are three reasons why AMD can't match Intel's profit ratios, or turn predictable profits every quarter.
1. AMD is spending twice as much (percentage-wise) as Intel on R&D. They've been playing catchup for years, and they still need spend a larger percentage than Intel just because most R&D costs don't escalate as your company's sales get larger.
2. AMD still only has a small piece of the server market. Although desktop sales heavily eclipse servers in numbers, the profit on all those overpriced server chips is incredible. Expect AMD's profits to become larger and more stable as they wrestle more share in this market from Intel.
3. Intel has stepped up their level of competitiveness in the flash market in the last year. It used to be AMD's flash business kept the x86 sector afloat, and now those tables have turned. There's not much you can do about this when you're 1/10 the size of your competition.
The 64-bit version of Far Cry was intended to give increased detail while still delivering similar performance to the 32-bit version.
There are more objects on maps, increased geometric complexity, improved effects, and the view distance has been increased. These are all typically CPU-limited performance barriers. See the [H]ardOCP review to get a feel for what 64-bit brings to the party. It's not an AMAZING improvement, but it is significant.
I'm just going to state for clarity that making the iMac "USB-only" was hardly "innovative".
Apple knew that there was a next-generation storage interface right around the corner...they just didn't know what would win out. Would we be using Zip drives, LS-120, Shark drives, Syquest SyJet discs...or could it be the many competing formats of writable / rewritable CD? Or could Sony pull off a surprise upset with a sudden marketing blitz pushing MiniDisc as a data drive?
Apple had no idea what media would win...so instead of taking a stance on any one, they sat back and watched. This isn't "innovating", it's called "following". By the time real next-generation storage standards emerged victorious (~2000 for the CD-R / RW, and ~2002 for the USB flash drive), the original concept of the iMac was long-gone, and every PC sold came with a CD writer and tons of USB ports.
I mean, when the iMac was released, what did users buy? USB Zip drives. USB floppy drives. No real innovation there, just a new connector.
You could argue that the innovation lay in Apple kick-starting the USB peripheral industry, except Apple did nothing of the sort. The iMac was painstakingly timed so it was released a half year before Windows 98. Microsoft had already been evangelizing the wonders of USB as a key selling feature, and had already been gearing up manufacturers to provide the backing.
Windows with full USB support was a HUGE market compared to the iMac...not only because every new PC would ship with 98 and plenty of USB ports, but because there was a huge upgrade market...millions of PCs sold in 1996 and 1997 with USB ports installed but limited software and peripheral support. And all those people would push and shove to be the first to buy USB gizmos.
About the only innovation the iMac can claim is the style...and one got the impression that they were the driving force behind USB peripherals because a number of early USB devices adopted the iMac look. But this was simply smart design on the part of the manufacturers; Steve Jobs was making as much noise as he could about the iMac, and the media couldn't get enough of it. Why not use the hype train to your advantage and build your perhiperals in a similar style?
This review is not applicable to the current discussion.
The discussion is about Firewire performance WITH RESPECT to the Ipod (in particular, the Nano with flash memory). Tom's review tests DESKTOP HARD DRIVES with an order of magnitude faster transfer rates than Nano.
The benchmarks in that article show that Firewire 400 has about a 10% lead over USB 2.0 for larger, faster drives, and about a 5% lead for slower drives. Obviously, it is the slightly increased access time for USB2 which hurts it in high-performance situations...but as maximum media transfer rates go down, the small increase in access time becomes insignificant.
Given that the Nano is a flash-based device, and couldn't hope to have a write speed faster than 4MB/s (there's no way they're offering higher-speed flash at those prices), there's little gained in offering Firewire.
This is the kind of thing USB2 was intended for. CHEAP, UNIVERSAL connection technology that is "good enough" for most cases. Firewire 400, as popular as it has become, still cannot offer even half the total marketshare USB can. And for a device like this, where the size of the board is the limiting factor (instead of the size of the drive on other iPods), each additional feature (chipset, busses and external connector) makes the board that much larger.
YES, Firewire 800 is freaking fast. NO, you don't need it unless you have devices on the bleeding-edge of performance. Not to mention you can hardly take advantage of it anywhere because only Powermacs and a handful of PCs support Firewire 800 speed.
You know, I can't believe how pointless top ten lists are, since it all comes doesn to "personal opinion."
Here is a short list of my favorite games:
Daikatana Battlecruiser 3000 AD Duke Nukem Forever
I don't know what the writers of this article were thinking. Star Craft made the list, yet Derek Smart's quintessential classic Batttlecruiser 3000 AD didn't even make a mention.
I know this is all personal opinion, but there's no way you can ignore John Romero's Daikatana and still call yourself an 'authority' on gaming.
with current housing prices going through the roof, it's financially prudent to rent as oppose to paying mortgage right now.
Especially if the place you live in has high costs of entry and high property taxes.
I'd love to buy a house, but instead I await the crash of the overinflated housing "market". I await the crash patiently while paying about $5000 per year in rent, sharing a house with 3 other people. Well...I could sign on for a $250,000 townhome, or a $350,000 single family home (this being the price range of the place I live in currently). For someone who has never been in the housing market, and lacks 2 incomes, it would be pure insanity to get in right now.
The sad fact is, thanks to the exhorbitant prices on homes here, my property taxes and home insurance costs would be almost as high as I pay in rent. I would still have to get roommates or a working wife to afford to buy the same level of house as I'm currently living in.
Furthermore, when you take into consideration all the invested time and money just to KEEP your equity (houses that fall apart do NOT increase in value), it's suddenly not all that enticing. My landlord pays for major repairs and major appliances. That would have been about $2000 out of my pocket over the last 2 years.
As with eating out... it's for my own sanity.. if i had to live on ramen after working 80-90 hours a week, my health care alone would exceed my cost of eating =)
Try working 40 hour weeks. I know, I know, it feels so unnatural...but give it a try sometime. You may find you have time enough to cook something healthy for yourself.
Right. S3 (owned by Via) was an obvious purchase back in 2000, because Via needed an integrated graphics unit to compete with Intel's integrated chipsets (Intel bought Real3D to carry this load for them, makers of the i740).
Via is taking a different route from Intel though, because they're not top-of-the-heap. Via, just like SiS (XGI), Nvidia and ATI, recognize that the only way to beat Intel is on features, because price for low-end chipsets is mostly flat.
So, even though Via has been milking the Savage architecture for their integrated chipset maeker, they know that eventually the DeltaChrome tech can be integrated in a chipset.
Of course, Via had better get their ass in gear. They've played all sorts of games with the Savage architecture (renaming it UniChrome to go with S3's new offerings), but they have yet to integrate a DX9 part. Neither has SiS. Meanwhile, Nvidia and ATI are bringing integrated x300SE / 6200TC parts to bear on Intel's pathetic Dx9 integrated solution. Looks like Via and SiS will be late to the party once again.
Agreed. I expect nothing from "new" graphics hardware from S3, because it alwasy invariably takes 2 years before hardware and driver tweaks get it up to promised performance.
In addition, S3 has always hindered their own hardware for strange reasons. The Savage 3D, released as a surprise comeback for S3 in 1998 as the first.25 micron video chipset, was pathetically limited to 8MB framebuffer. S3 sold the world on S3 Texture Compression mostly as a gimmick to avoid putting 16MB ram on their cards.
In addition, the Savage 3D could never quite keep up with the competition due to a restrictive 64-bit memory interface. S3 decided to shoot themselves in the foot again in early 1999, when they released the Savage 4 with a second pixel pipeline...and the same pathetic 64-bit memory interface. Whereas the Savage 3D could get away with 16-bit color, this was 1999, and 32-bit color performance was key. The Savage 4 hit a severe memory bottneck in resolutions 800x600x32 and above.
Then, late in 1999, S3 releases the Savage 2000 in response to the GeForce 256. The performance was surprisingly good in OpenGL, but the Direct3D performance sucked, and the promised T&L unit turned out to be non-functional. Too bad, S3 is too late to the party. The release of the GeForce 2 GTS early next spring smashed any hope of S3 pulling off a winner there.
There will be a 33-50% performance hit for OpenGL on Vista, and only up to OpenGL 1.4 will be supported, without extensions. Unline previous versions of Windows, there will be no way for video card vendors to provide ICDs.
What does this mean in the long run? I imagine even ID Software will have to give in and develop their next game in Direct 3D.
I'm just glad it never made it past the initial adoption for workstations phase. For such an "elegant" design, it needs a brutal amount of L2 and L3 cache.
When you see what an Athlon 64 can do with a paltry 512k L2 cache, you get a real appreciation for instruction reordering, register renaming and branch prediction.
Intel says: "If you buy AMD chips for your notebooks you'll just get a chip. You'll still have to go and part out motherboards, wireless chips, video cards, everything. On the other hand, you go with Intel, we'll give you an integrated motherboard with everything. You get the whole ball of wax from us, cheap! And if you want, you can pay us for a license to the Centrino brand name, too."
That's an incentive to buy Intel. It's also a disincentive against buying AMD. It also sounds like good business sense to me, not any kind of crime
Except that this is not quite the case. Intel's Centrino platform is not completely integrated: it uses a MiniPCI card for wireless, and is required for a platform to carry the Centrino name. The catch? The ONLY MiniPCI card allowed is the Intel Pro/Wireless card.
Manufacturers were very pissed off that they couldn't offer a Pentium M system without wireless and still sell it under Centrino, let alone provide a different wireless card.
You could get exactly the same level of integration on the AMD side for years (VIA and SiS offered integrated video chipsets for years) for less, with the possibility of including an even better wireless MiniPCI card. But Intel plastered the Centrino name all over the media, basically telling manufacturers that if they ignored Centrino, they would be left behind.
They are opening the new Fab 36 in 2005/06, which should roughly double AMD's production capacity (their two current 8-inch fabs produce less than the new 12-inch fab can).
Apple's sales in 2004 were about 3.3 million computers.
So, you do the math: roughly 5-10% of AMD's capacity (depending on how troublesome Fab 36 is) is a pretty big drop in the bucket, especially since AMD is currently so stretched to meet their current supplier's demands that they're outsourcing chip production.
But, for Intel, who should increase capacity by a huge jump in the next year, Apple is no strain.
Funny you should mention Tritech, they bought the design for the Pyramid from Bitboys Oy.
The design was advanced, but in the marketing department they were idiots. They actually did fab one of their designs in 1997, but few board builders bought it because they stupidly did not include a VGA core.
Back in 1996, when 2D cards had either no 3D features, or had limited "FreeD" features that were cheap to bundle into existing designs (ATI Rage IIc, S3 ViRGE, Matrox Millenium s220), standalone accelerators like the PowerVR PCX1 and Voodoo Graphics could still make headway.
But this was 1997, and the market was getting decidedly polarized toward combo cards. The ATI Rage Pro, Rendition v2200, Nvidia Riva 128, Oak Warp 5, Number 9 Revolution, Permedia 2 and even the 3dfx Voodo Rush were making performance integrated 2D/3D a "must have".
So, Tritech wasted all their money on a 3D-only design, Nobody bought into it., and they never had the chance to fab their 2D/3D combo chipset.
Rendition (Made the Verite, featured FAST edge AA. Ran the first 3D-accelerated port of Quake. Absorbed by Micron) Oak Technologies (made the Warp 5, a tile-based chip much like PowerVR, which featured FSAA) Real3D (designed the i740 for Intel, intended to make AGP THE graphics standard. They were probably later absorbed by Intel) 3DFX (One word: Glide. Now part of Nvidia)
Oh, and the Kyro chip was made by ST Micro, with IP licensed from NEC / PowerVR.
Matrox is basically down for the count. They will NEVER catch up, not now. Matrox caught up... ONCE, with the G400. The collossal failures of the Parhelia is a testament to the fact that they're long gone from the performance arena.
Matrox will bleed the G line to death until they can no longer compete on price or features, then continue to offer custom video hardware solutions while they shop for a buyer.
I see you forgot to mention S3 in that set. S3 is alive and...well, doing OK.
VIA knows that the most money to be made is in embedded IP, that's why they bought S3 to get the Savage 3D now used in their integrated video chipsets. VIA launched the DeltaChrome for the future, even though it sucks in today's market. VIA knows that the DeltaChrome cannot compete now, but in a few years when the drivers are finally mature it can be slimmed down and replace the aging Savage core.
Where once we had performance as king, now savvy markinging and lowest price is king.
I went to Puerto Rico over the summer, and one interesting aspect of their architecture is the prevalance of stone and metal structure, especially near the coasts.
Their designs typically have no glass (unless you're in touristy areas, which are more westernized)...windows are covered by metal slats which can be closed when the weather gets bad. Entire living areas of houses are open air, with bars to keep out intruders.
We could learn a thing or two from these folks, if we want to build in hurricane country for the long term.
Freelancer was a marginal game in single-player, but the persistent universe server online play made it endearing.
Servers sprang up with all kinds of concepts, from user-enforced rules (RP) to complete mods with entirely new universes.
The Freelancer universe was a solid foundation, and the flight engine was excellent for PvP combat. Go have a look at the online community that still persists after nearly 3 years, if you need a second opinion.
But this patent isn't like a poison or drug effect, it was filed in response to Eternal Darkness.
This game is UNIQUE. You have a sanity meter, and as it goes down your character starts hallucinating. You bring the meter back up by killing enemies, and thus regaining your confidence. Sometimes, the hallucination was obvious, designed to make you laugh (walls bleeding, walking on the celing, strange noises, etc).
But some instances were devilishly clever. One time, I was playing late at night, with the lights off. Suddenly, the sound cuts out and I see a big pixelated "MUTE" on the screen.
I start looking around in the dark, trying to see if my stupid ass had rolled over the remote, when the sound suddenly cut in and my character screamed "WHAT IS GOING ON!". Freaked me out.
It doesn't matter that the "MUTE" didn't look quite like my TV's overlay, at that point I was too into the game to think that out. Best trick ever pulled on a player. Why is this unique? The nastier tricks were rare, and never repeated (something you can't say for, say, status ailment effects, which are usually the same, or predictable).
Other nasty tricks that only happened once:
Hallucinating and seeing additional ghouls in an area I'd already cleared, with them appearing right behind me.
Hallucinating that I'd blown my head off trying to reload a flintlock pistol. Thought I'd have to restart the whole battle until the hallucination ended.
There's the problem in a nutshell - the certification is a way for competing authorities to milk money out of people, and worse, if the model takes off amongst employers, then it will become a compulsory way to milk money out of people because independent learning will not be recognized.
I thought they called that College.
Just try getting past the HR monkey without at least a BS, I dare ya.
Diamond Multimedia was on the road to recovery. Despite falling sales of modems, and the video card market crash after 3dfx started making their own boards (which also claimed Jazz, Hercules, Orchid and Canopus's US market, just to name a few), DMM had made smart moves into selling motherboards and, of course, the Rio.
Then Diamond made the boneheaded decision to purchase S3. It was like they had done a complete 180...S3 was in serious trouble, and Diamond was in no position to bail them out.
The Rio's successor was more of the same: just more built-in memory, no new features. As a result, they lost momentuum.
Eventually, Diamond faltered under the wave of crap. S3 was sold off to VIA, and the audio division of Diamond became SonicBlue. Then ReplayTV sucked, and SonicBlue missed the boat on small hard drive mp3 players.
So, you see the lovely lack of foundation SonicBlue has been trying to stand on. I wonder what they're going to dop now that they sell virtually nothing. Maybe sell off the name to some other company.
Indicentally, I've noticed that the Diamond name has been revived recently, not a bad move for a distributer wanting to open a new market in the US.
(arguably) the era of Street Fighter2 wiping everything else out of the arcades. The fighting games chased away all but the diehards and subsequently outside of Japan, the arcades died.
I'll argue that.
Arcades were already dying in the US during the 1980s due to forces completely beyond their control. Arcades were already slated to die for the same reason bowling alleys, shopping malls and downtown areas have seen declines (just to name a few). Entertainment, shopping, you name it...it has call become detached and depersonalized.
Of course, the gaming industry didn't help themselves. The console wars of the 1970s started the wave: you may not think the Atari VCS has good graphics, but on release in 1977 they were as good as ANY arcade machine. Why the hell would people DRIVE to an arcade when they could have the arcade at home? The video game market crash of 1984 didn't help things, and the renewed console wars of the late 80s sealed the deal.
The fighting game was a last gasp to make gaming in the arcades "social" again, by adding so much complexity to a game and practically forcing people to face off rather than settle for the "AI". It worked for awhile, with dozens of knockoffs, each with new killer combos to learn. However, it was not enough to save the industry in America...there are still "arcades" here in the US, but they are usually a secondary part of another business.
Slashdot Mirror
I have to agree, Via Mini-ITX are a lackluster choice for desktop use.
The problem is the anemic processor bus, which makes any rich content seem sluggish. Thanks to the Nemiah core upgrade, the processor is perfectly capable, but any modern GUI interface (XP, GNOME, KDE) will feel slow. Overall performance under load also suffers, as it becomes bus-bound.
An upgrade to a 200MHz revamp of the GTL bus and a slightly larger L2 cache would make this chip competitve without sacrificing cost or performance. Now especially that Via is offering a combined C3 core and northbridge on a single package, there's no reason the bus couldn't be pumped up (similar to what Freescale did with their integrated 667MHz MPX bus on the dual core MPC8641D).
Unfortunately for us, the Via Epia is already overpowered for it's real market: kiosk, POS, and other low-power specialty applications benefiting from x86 compatibility. Via is happy to release variants with LVDS panel controllers and compactflash slots, but there is no market need for improving processor bandwidth.
So, get used to that poor desktop performance. Luckily, AMD and Intel are starting to intrude on Via's low-power computer enthusiast market with the Athlon 64 and the Pentium M.
Actually, this has more to do with Intel pushing a . They expect leakage to be reduced by over 100 times.
Basically, the article is simply parroting what the Intel marketing department has been told by their engineers: that leakage current will be reduced by allmost 1000 times. The Intel marketing department is simply palming it off as some newfangled "ultra low power" process, when it's been on the roadmap the whole time.
That is, if the reality about high-K comes anywhere near Intel's claims. I recall quite clearly Intel skipping out on SOI, and claiming quite calmly to the world that strained silicon would solve all their problems in the move to 90nm. That didn't happen. Ultra-aggressive dynamic power-saving modes and the Pentium M are the only things that saved Intel's 90nm process.
Will be impressive if they actually get results in the same ballpark as their estimates, though.
I'd like to add that the Turion ML processor used in that test is part of the higher-power usage cores with a TDP of 35w.
The Turion MT processor (25w TDP) is available all the way up to 2.2GHz (since the end of August), and is QUITE competitve in terms of power usage with the Pentium M, even at full load.
They're evenly-matched in the innovation arena. If you folks really want to know why Apple chose Intel over AMD, I've already run the numbers. Though AMD is quire innovative and competitive, it would be risky for Apple to depend solely on AMD.
My system is faster than hers at integer operations but throw something like world of warcraft at it and hers is faster
she now has 1.25gb ram. (radeon 9800 128mb agp 4x vs my 9600 aiw xt 128mb agp 8x)
Yeah, you get back to me when the 9600 suddenly performs anywhere near as well as a 9800.
If you were trying to make a point, perhaps you'd better start over from the beginning.
And? There are three reasons why AMD can't match Intel's profit ratios, or turn predictable profits every quarter.
1. AMD is spending twice as much (percentage-wise) as Intel on R&D. They've been playing catchup for years, and they still need spend a larger percentage than Intel just because most R&D costs don't escalate as your company's sales get larger.
2. AMD still only has a small piece of the server market. Although desktop sales heavily eclipse servers in numbers, the profit on all those overpriced server chips is incredible. Expect AMD's profits to become larger and more stable as they wrestle more share in this market from Intel.
3. Intel has stepped up their level of competitiveness in the flash market in the last year. It used to be AMD's flash business kept the x86 sector afloat, and now those tables have turned. There's not much you can do about this when you're 1/10 the size of your competition.
The 64-bit version of Far Cry was intended to give increased detail while still delivering similar performance to the 32-bit version.
There are more objects on maps, increased geometric complexity, improved effects, and the view distance has been increased. These are all typically CPU-limited performance barriers. See the [H]ardOCP review to get a feel for what 64-bit brings to the party. It's not an AMAZING improvement, but it is significant.
I'm just going to state for clarity that making the iMac "USB-only" was hardly "innovative".
Apple knew that there was a next-generation storage interface right around the corner...they just didn't know what would win out. Would we be using Zip drives, LS-120, Shark drives, Syquest SyJet discs...or could it be the many competing formats of writable / rewritable CD? Or could Sony pull off a surprise upset with a sudden marketing blitz pushing MiniDisc as a data drive?
Apple had no idea what media would win...so instead of taking a stance on any one, they sat back and watched. This isn't "innovating", it's called "following". By the time real next-generation storage standards emerged victorious (~2000 for the CD-R / RW, and ~2002 for the USB flash drive), the original concept of the iMac was long-gone, and every PC sold came with a CD writer and tons of USB ports.
I mean, when the iMac was released, what did users buy? USB Zip drives. USB floppy drives. No real innovation there, just a new connector.
You could argue that the innovation lay in Apple kick-starting the USB peripheral industry, except Apple did nothing of the sort. The iMac was painstakingly timed so it was released a half year before Windows 98. Microsoft had already been evangelizing the wonders of USB as a key selling feature, and had already been gearing up manufacturers to provide the backing.
Windows with full USB support was a HUGE market compared to the iMac...not only because every new PC would ship with 98 and plenty of USB ports, but because there was a huge upgrade market...millions of PCs sold in 1996 and 1997 with USB ports installed but limited software and peripheral support. And all those people would push and shove to be the first to buy USB gizmos.
About the only innovation the iMac can claim is the style...and one got the impression that they were the driving force behind USB peripherals because a number of early USB devices adopted the iMac look. But this was simply smart design on the part of the manufacturers; Steve Jobs was making as much noise as he could about the iMac, and the media couldn't get enough of it. Why not use the hype train to your advantage and build your perhiperals in a similar style?
This review is not applicable to the current discussion.
The discussion is about Firewire performance WITH RESPECT to the Ipod (in particular, the Nano with flash memory). Tom's review tests DESKTOP HARD DRIVES with an order of magnitude faster transfer rates than Nano.
The benchmarks in that article show that Firewire 400 has about a 10% lead over USB 2.0 for larger, faster drives, and about a 5% lead for slower drives. Obviously, it is the slightly increased access time for USB2 which hurts it in high-performance situations...but as maximum media transfer rates go down, the small increase in access time becomes insignificant.
Given that the Nano is a flash-based device, and couldn't hope to have a write speed faster than 4MB/s (there's no way they're offering higher-speed flash at those prices), there's little gained in offering Firewire.
This is the kind of thing USB2 was intended for. CHEAP, UNIVERSAL connection technology that is "good enough" for most cases. Firewire 400, as popular as it has become, still cannot offer even half the total marketshare USB can. And for a device like this, where the size of the board is the limiting factor (instead of the size of the drive on other iPods), each additional feature (chipset, busses and external connector) makes the board that much larger.
YES, Firewire 800 is freaking fast. NO, you don't need it unless you have devices on the bleeding-edge of performance. Not to mention you can hardly take advantage of it anywhere because only Powermacs and a handful of PCs support Firewire 800 speed.
You know, I can't believe how pointless top ten lists are, since it all comes doesn to "personal opinion."
Here is a short list of my favorite games:
Daikatana
Battlecruiser 3000 AD
Duke Nukem Forever
I don't know what the writers of this article were thinking. Star Craft made the list, yet Derek Smart's quintessential classic Batttlecruiser 3000 AD didn't even make a mention.
I know this is all personal opinion, but there's no way you can ignore John Romero's Daikatana and still call yourself an 'authority' on gaming.
with current housing prices going through the roof, it's financially prudent to rent as oppose to paying mortgage right now.
... it's for my own sanity.. if i had to live on ramen after working 80-90 hours a week, my health care alone would exceed my cost of eating =)
Especially if the place you live in has high costs of entry and high property taxes.
I'd love to buy a house, but instead I await the crash of the overinflated housing "market". I await the crash patiently while paying about $5000 per year in rent, sharing a house with 3 other people. Well...I could sign on for a $250,000 townhome, or a $350,000 single family home (this being the price range of the place I live in currently). For someone who has never been in the housing market, and lacks 2 incomes, it would be pure insanity to get in right now.
The sad fact is, thanks to the exhorbitant prices on homes here, my property taxes and home insurance costs would be almost as high as I pay in rent. I would still have to get roommates or a working wife to afford to buy the same level of house as I'm currently living in.
Furthermore, when you take into consideration all the invested time and money just to KEEP your equity (houses that fall apart do NOT increase in value), it's suddenly not all that enticing. My landlord pays for major repairs and major appliances. That would have been about $2000 out of my pocket over the last 2 years.
As with eating out
Try working 40 hour weeks. I know, I know, it feels so unnatural...but give it a try sometime. You may find you have time enough to cook something healthy for yourself.
Right. S3 (owned by Via) was an obvious purchase back in 2000, because Via needed an integrated graphics unit to compete with Intel's integrated chipsets (Intel bought Real3D to carry this load for them, makers of the i740).
Via is taking a different route from Intel though, because they're not top-of-the-heap. Via, just like SiS (XGI), Nvidia and ATI, recognize that the only way to beat Intel is on features, because price for low-end chipsets is mostly flat.
So, even though Via has been milking the Savage architecture for their integrated chipset maeker, they know that eventually the DeltaChrome tech can be integrated in a chipset.
Of course, Via had better get their ass in gear. They've played all sorts of games with the Savage architecture (renaming it UniChrome to go with S3's new offerings), but they have yet to integrate a DX9 part. Neither has SiS. Meanwhile, Nvidia and ATI are bringing integrated x300SE / 6200TC parts to bear on Intel's pathetic Dx9 integrated solution. Looks like Via and SiS will be late to the party once again.
Agreed. I expect nothing from "new" graphics hardware from S3, because it alwasy invariably takes 2 years before hardware and driver tweaks get it up to promised performance.
.25 micron video chipset, was pathetically limited to 8MB framebuffer. S3 sold the world on S3 Texture Compression mostly as a gimmick to avoid putting 16MB ram on their cards.
In addition, S3 has always hindered their own hardware for strange reasons. The Savage 3D, released as a surprise comeback for S3 in 1998 as the first
In addition, the Savage 3D could never quite keep up with the competition due to a restrictive 64-bit memory interface. S3 decided to shoot themselves in the foot again in early 1999, when they released the Savage 4 with a second pixel pipeline...and the same pathetic 64-bit memory interface. Whereas the Savage 3D could get away with 16-bit color, this was 1999, and 32-bit color performance was key. The Savage 4 hit a severe memory bottneck in resolutions 800x600x32 and above.
Then, late in 1999, S3 releases the Savage 2000 in response to the GeForce 256. The performance was surprisingly good in OpenGL, but the Direct3D performance sucked, and the promised T&L unit turned out to be non-functional. Too bad, S3 is too late to the party. The release of the GeForce 2 GTS early next spring smashed any hope of S3 pulling off a winner there.
Unfortunately, thanks to Microsoft permanantly degrading OpenGL perfornance in Vista, it would be foolhardy of ANY developer to choose SDL / OpenGL at this time.
There will be a 33-50% performance hit for OpenGL on Vista, and only up to OpenGL 1.4 will be supported, without extensions. Unline previous versions of Windows, there will be no way for video card vendors to provide ICDs.
What does this mean in the long run? I imagine even ID Software will have to give in and develop their next game in Direct 3D.
I'm just glad it never made it past the initial adoption for workstations phase. For such an "elegant" design, it needs a brutal amount of L2 and L3 cache.
When you see what an Athlon 64 can do with a paltry 512k L2 cache, you get a real appreciation for instruction reordering, register renaming and branch prediction.
Intel says: "If you buy AMD chips for your notebooks you'll just get a chip. You'll still have to go and part out motherboards, wireless chips, video cards, everything. On the other hand, you go with Intel, we'll give you an integrated motherboard with everything. You get the whole ball of wax from us, cheap! And if you want, you can pay us for a license to the Centrino brand name, too."
That's an incentive to buy Intel. It's also a disincentive against buying AMD. It also sounds like good business sense to me, not any kind of crime
Except that this is not quite the case. Intel's Centrino platform is not completely integrated: it uses a MiniPCI card for wireless, and is required for a platform to carry the Centrino name. The catch? The ONLY MiniPCI card allowed is the Intel Pro/Wireless card.
Manufacturers were very pissed off that they couldn't offer a Pentium M system without
wireless and still sell it under Centrino, let alone provide a different wireless card.
You could get exactly the same level of integration on the AMD side for years (VIA and SiS offered integrated video chipsets for years) for less, with the possibility of including an even better wireless MiniPCI card. But Intel plastered the Centrino name all over the media, basically telling manufacturers that if they ignored Centrino, they would be left behind.
It's not illegal, but it is very underhanded.
It's not easy to find raw sales numbers, but here goes:
AMD sold 36 million processors in 2004.
They are opening the new Fab 36 in 2005/06, which should roughly double AMD's production capacity (their two current 8-inch fabs produce less than the new 12-inch fab can).
Intel should have 375 million per year capacity in 2005/06, thanks to 5 new 12-inch fabs.
Apple's sales in 2004 were about 3.3 million computers.
So, you do the math: roughly 5-10% of AMD's capacity (depending on how troublesome Fab 36 is) is a pretty big drop in the bucket, especially since AMD is currently so stretched to meet their current supplier's demands that they're outsourcing chip production.
But, for Intel, who should increase capacity by a huge jump in the next year, Apple is no strain.
Funny you should mention Tritech, they bought the design for the Pyramid from Bitboys Oy.
The design was advanced, but in the marketing department they were idiots. They actually did fab one of their designs in 1997, but few board builders bought it because they stupidly did not include a VGA core.
Back in 1996, when 2D cards had either no 3D features, or had limited "FreeD" features that were cheap to bundle into existing designs (ATI Rage IIc, S3 ViRGE, Matrox Millenium s220), standalone accelerators like the PowerVR PCX1 and Voodoo Graphics could still make headway.
But this was 1997, and the market was getting decidedly polarized toward combo cards. The ATI Rage Pro, Rendition v2200, Nvidia Riva 128, Oak Warp 5, Number 9 Revolution, Permedia 2 and even the 3dfx Voodo Rush were making performance integrated 2D/3D a "must have".
So, Tritech wasted all their money on a 3D-only design, Nobody bought into it., and they never had the chance to fab their 2D/3D combo chipset.
Rendition (Made the Verite, featured FAST edge AA. Ran the first 3D-accelerated port of Quake. Absorbed by Micron)
... ONCE, with the G400. The collossal failures of the Parhelia is a testament to the fact that they're long gone from the performance arena.
Oak Technologies (made the Warp 5, a tile-based chip much like PowerVR, which featured FSAA)
Real3D (designed the i740 for Intel, intended to make AGP THE graphics standard. They were probably later absorbed by Intel)
3DFX (One word: Glide. Now part of Nvidia)
Oh, and the Kyro chip was made by ST Micro, with IP licensed from NEC / PowerVR.
Matrox is basically down for the count. They will NEVER catch up, not now. Matrox caught up
Matrox will bleed the G line to death until they can no longer compete on price or features, then continue to offer custom video hardware solutions while they shop for a buyer.
I see you forgot to mention S3 in that set. S3 is alive and...well, doing OK.
VIA knows that the most money to be made is in embedded IP, that's why they bought S3 to get the Savage 3D now used in their integrated video chipsets. VIA launched the DeltaChrome for the future, even though it sucks in today's market. VIA knows that the DeltaChrome cannot compete now, but in a few years when the drivers are finally mature it can be slimmed down and replace the aging Savage core.
Where once we had performance as king, now savvy markinging and lowest price is king.
Exactly.
I went to Puerto Rico over the summer, and one interesting aspect of their architecture is the prevalance of stone and metal structure, especially near the coasts.
Their designs typically have no glass (unless you're in touristy areas, which are more westernized)...windows are covered by metal slats which can be closed when the weather gets bad. Entire living areas of houses are open air, with bars to keep out intruders.
We could learn a thing or two from these folks, if we want to build in hurricane country for the long term.
Bull.
Freelancer was a marginal game in single-player, but the persistent universe server online play made it endearing.
Servers sprang up with all kinds of concepts, from user-enforced rules (RP) to complete mods with entirely new universes.
The Freelancer universe was a solid foundation, and the flight engine was excellent for PvP combat. Go have a look at the online community that still persists after nearly 3 years, if you need a second opinion.
But this patent isn't like a poison or drug effect, it was filed in response to Eternal Darkness.
This game is UNIQUE. You have a sanity meter, and as it goes down your character starts hallucinating. You bring the meter back up by killing enemies, and thus regaining your confidence. Sometimes, the hallucination was obvious, designed to make you laugh (walls bleeding, walking on the celing, strange noises, etc).
But some instances were devilishly clever. One time, I was playing late at night, with the lights off. Suddenly, the sound cuts out and I see a big pixelated "MUTE" on the screen.
I start looking around in the dark, trying to see if my stupid ass had rolled over the remote, when the sound suddenly cut in and my character screamed "WHAT IS GOING ON!". Freaked me out.
It doesn't matter that the "MUTE" didn't look quite like my TV's overlay, at that point I was too into the game to think that out. Best trick ever pulled on a player. Why is this unique? The nastier tricks were rare, and never repeated (something you can't say for, say, status ailment effects, which are usually the same, or predictable).
Other nasty tricks that only happened once:
Hallucinating and seeing additional ghouls in an area I'd already cleared, with them appearing right behind me.
Hallucinating that I'd blown my head off trying to reload a flintlock pistol. Thought I'd have to restart the whole battle until the hallucination ended.
There's the problem in a nutshell - the certification is a way for competing authorities to milk money out of people, and worse, if the model takes off amongst employers, then it will become a compulsory way to milk money out of people because independent learning will not be recognized.
I thought they called that College.
Just try getting past the HR monkey without at least a BS, I dare ya.
It's more like this:
Diamond Multimedia was on the road to recovery. Despite falling sales of modems, and the video card market crash after 3dfx started making their own boards (which also claimed Jazz, Hercules, Orchid and Canopus's US market, just to name a few), DMM had made smart moves into selling motherboards and, of course, the Rio.
Then Diamond made the boneheaded decision to purchase S3. It was like they had done a complete 180...S3 was in serious trouble, and Diamond was in no position to bail them out.
The Rio's successor was more of the same: just more built-in memory, no new features. As a result, they lost momentuum.
Eventually, Diamond faltered under the wave of crap. S3 was sold off to VIA, and the audio division of Diamond became SonicBlue. Then ReplayTV sucked, and SonicBlue missed the boat on small hard drive mp3 players.
So, you see the lovely lack of foundation SonicBlue has been trying to stand on. I wonder what they're going to dop now that they sell virtually nothing. Maybe sell off the name to some other company.
Indicentally, I've noticed that the Diamond name has been revived recently, not a bad move for a distributer wanting to open a new market in the US.
Yes, you can find the Magic 8 Ball in the pool table in Bishop's House.
In addition to the usual snide responses, it also tells you useful information for key plot points, plus where to get goodies.
But, there are lots of other fun easter eggs in Fallout 2...all you need is excellent luck and high outdoorsman skill to encounter them.
(arguably) the era of Street Fighter2 wiping everything else out of the arcades. The fighting games chased away all but the diehards and subsequently outside of Japan, the arcades died.
I'll argue that.
Arcades were already dying in the US during the 1980s due to forces completely beyond their control. Arcades were already slated to die for the same reason bowling alleys, shopping malls and downtown areas have seen declines (just to name a few). Entertainment, shopping, you name it...it has call become detached and depersonalized.
Of course, the gaming industry didn't help themselves. The console wars of the 1970s started the wave: you may not think the Atari VCS has good graphics, but on release in 1977 they were as good as ANY arcade machine. Why the hell would people DRIVE to an arcade when they could have the arcade at home? The video game market crash of 1984 didn't help things, and the renewed console wars of the late 80s sealed the deal.
The fighting game was a last gasp to make gaming in the arcades "social" again, by adding so much complexity to a game and practically forcing people to face off rather than settle for the "AI". It worked for awhile, with dozens of knockoffs, each with new killer combos to learn. However, it was not enough to save the industry in America...there are still "arcades" here in the US, but they are usually a secondary part of another business.