Domain: endian.net
Stories and comments across the archive that link to endian.net.
Comments · 29
-
Nitpick/Warning: Radeon 9100 is not DirectX 9
I have a motherboard with an ATI Radeon 9100 IGP chipset in one of my machines at the office - the video chipset is a pig... It's a DirectX 9 compatible chipset
I'm not disagreeing with what I think is your comment's point, but the Radeon 9100 is not DirectX 9 compatible. It will not meet Vista's DX9 requirement. ATI's DX9 GPUs start at Radeon 9500 (AGP) and X300 (PCIe). The deceptively named Radeon 9000-9200 GPUs, which were released in the same generation as 9500-9800, are based on the Radeon 8500, a DX8 part. Here's endian.net's page on Radeon 8500/9100:endian.net: ATI Radeon 8500/9100
Add to that your motherboard's shared system and graphics memory bandwidth, it's no surprise that the integrated Radeon 9100 lags behind the a Radeon 8500 (without shared memory) and a GeForce FX 5200 (a true DX9 card). The Radeon 7500 is a little surprising, though. Did you configure your shared system/graphics memory in dual-channel mode? -
Re:Energy efficiency
[quote] No - Sun manage to get four multithreading cores in their Niagra, and only run at 72 watts with 32 threads. see this However, with Intel's cores, I expect be able to have a hot dinner faster than you can say Microwave". [/quote]
Oh, do you expect to fry an egg with a 31W (TDP) Sossaman, or do you expect to do it with a (TDP not confirmed officially) 80W Woodcrest? I can understand faboi comments like these, but for it to get modded to 4-Interesting is seriously lame. -
Re:comparison against the G4?For PowerMac G5 users, no way. The intel is a 32-bit chip which would likely limit me to 2GB of ram.
Just a nitpick: Core Duo's 32-bitness doesn't limit you to 2GB of RAM. Without getting technical:
2^32 = 4,294,967,296
Intel's 945 chipset for Core Duo (which the MacBook Pro and iMac uses) supports up to 4GB.The PowerMac will undoubtedly wait for the 64-bit Conroe desktop processors or Woodcrest workstation/server processors, which will arrive in July at the earliest.
-
Re:comparison against the G4?For PowerMac G5 users, no way. The intel is a 32-bit chip which would likely limit me to 2GB of ram.
Just a nitpick: Core Duo's 32-bitness doesn't limit you to 2GB of RAM. Without getting technical:
2^32 = 4,294,967,296
Intel's 945 chipset for Core Duo (which the MacBook Pro and iMac uses) supports up to 4GB.The PowerMac will undoubtedly wait for the 64-bit Conroe desktop processors or Woodcrest workstation/server processors, which will arrive in July at the earliest.
-
Re:Bandwidth
I wonder whether the quad-core Intel chips will be as bandwidth-starved as the dual-core chips? Currently, the comparison between a dual-core Pentium and a dual-core Opteron is farcical, especially for memory-limited apps.
The quad-core chip that Intel demoed in TFA, code name Cloverton, is derived from the Merom/Conroe/Woodcrest architecture (which is derived from Yonah/Core Duo). It is not derived from the current bandwidth-starved Xeon core (Netburst/Pentium 4).I agree that the comparison between dual-core Pentium D/Xeon and dual-core Athlon 64/Opteron is a farce. However, Yonah (even with its 667MHz bus and non-integrated memory controller) did quite well against Athlon 64.
Of course, that doesn't mean bandwidth won't be a problem for Cloverton when they move to 4 cores. I believe Cloverton will still have a FSB and seperate memory controller while Opteron will have next-gen HyperTransport and integrated memory controller(s).
-
Re:Actually, they were outdone..
AMD made this very same announcement in June/2005, indicating quad-core CPUs would be available 'sometime in 2007'.
But have they demonstrated working quad-core CPUs the way Intel did in TFA? I think an "announcement" is more impressive if they can demonstrate at least a working prototype. From TFA:To show the product is well on its way, Intel CTO Justin Rattner demonstrated a working server computer with a pair of the new microprocessors, code-named Clovertown.
BTW, Cloverton is based on the Merom/Conroe/Woodcrest architecture, not on the current Netburst/Pentium 4 architecture. -
Re:Pro appsI found it most disapointing that Apple decided to go from the G5 64-bit chip to a Core Duo which is a double die 32-bit chip.
I think everybody was surprised when the G5 iMac switched to a 32-bit chip. Who predicted it? All predictions I saw only mentioned the G4 Macs (mini, iBook, PowerBook) as possible candidates for 32-bit Yonah.
Seems bass ackwards to me.
After the initial shock, it actually seems reasonable to me. Despite its 32-bitness, Core Duo is very fast, cool (designed for notebooks), and dual-core. Intel's switch to 65nm seems to be going much smoother than their (and IBM's) switch to 90nm. Is IBM producing a fast dual-core G5 that's cool enough for the iMac's form factor and cheap enough for the iMac's price point? Remember, Apple touted "performance per watt" when explaining the switch to Intel.
It seems that that would only hurt performance in "pro" lever apps. But hey I guess we'llhave to wait and see what the towers have in them before passing judgement.
The towers will have Conroe or Woodcrest, both of which are 64-bit CPUs and are due in July at the earliest. The iMac will also probably move to Conroe later this year.
I'm not convinced that, over a 5-7 year lifetime of an iMac, a reduced-power single-core G5 iMac would outperform a Core Duo iMac in pro apps. First, the iMac's form factor probably limits it to 2GB of memory. Given this limitation, will 64-bit pro apps on reduced-speed iMac G5 perform better than the (eventually) optimized 32-bit versions on the iMac Core Duo?
For the iMac's form factor, I think the Core Duo's high performance per watt and dual cores will make up for its limitations for being 32-bit.
-
Re:Pro appsI found it most disapointing that Apple decided to go from the G5 64-bit chip to a Core Duo which is a double die 32-bit chip.
I think everybody was surprised when the G5 iMac switched to a 32-bit chip. Who predicted it? All predictions I saw only mentioned the G4 Macs (mini, iBook, PowerBook) as possible candidates for 32-bit Yonah.
Seems bass ackwards to me.
After the initial shock, it actually seems reasonable to me. Despite its 32-bitness, Core Duo is very fast, cool (designed for notebooks), and dual-core. Intel's switch to 65nm seems to be going much smoother than their (and IBM's) switch to 90nm. Is IBM producing a fast dual-core G5 that's cool enough for the iMac's form factor and cheap enough for the iMac's price point? Remember, Apple touted "performance per watt" when explaining the switch to Intel.
It seems that that would only hurt performance in "pro" lever apps. But hey I guess we'llhave to wait and see what the towers have in them before passing judgement.
The towers will have Conroe or Woodcrest, both of which are 64-bit CPUs and are due in July at the earliest. The iMac will also probably move to Conroe later this year.
I'm not convinced that, over a 5-7 year lifetime of an iMac, a reduced-power single-core G5 iMac would outperform a Core Duo iMac in pro apps. First, the iMac's form factor probably limits it to 2GB of memory. Given this limitation, will 64-bit pro apps on reduced-speed iMac G5 perform better than the (eventually) optimized 32-bit versions on the iMac Core Duo?
For the iMac's form factor, I think the Core Duo's high performance per watt and dual cores will make up for its limitations for being 32-bit.
-
Re:WowI really have to wonder when Intel will start using this technology in desktops.
Not in a big way until Conroe in the second half of 2006. I'm pretty sure "desktop" use of Yonah will be limited to a few small form factor desktops like Dothan is used today.
Yonah will still be 32-bit while Intel's entire "desktop" line of CPUs (including Celeron) have adopted EM64T. I don't think this is that important, but Conroe will add EM64T and other enhancements to the "Pentium M core."
65nm Pentium 4/D processors (Cedar Mill/Presler) are launching around the same time as Yonah (January) and are already shipping in volume to manufacturers. Another Anandtech article showed significant power savings from 65nm, which will make Pentium 4/D power consumption reasonable (but not quite as low as AMD 90nm). In addition, Pentium 4/D will also add virtualization technology (Vanderpool) which Yonah will lack.
Intell is should switch its R&D and support the Pentium M as a desktop chip
I don't think the Pentium M architecture is lacking in R&D (see Conroe). Yonah is making some steps toward being a desktop chip by adding SSE3 and floating point enhancements, but Conroe will complete the transition by adding EM64T, virtualization technology, desktop chipsets, and more lenient power requirements.
-
Re:ibook vs. powerbookI highly doubt that. The Yonah-cores are due january, and I doubt if Intel will ship it as Celeron first. It's more likely that they ship it as Pentium M with the Celeron M following a few months later: too late for the introduction of a Celeron-based iBook in january.
I think he means the current Celeron M. I don't know why everybody is assuming Apple will use the upcoming 65nm Yonah core so soon.
A current Dothan-based Celeron M (Pentium M's core, 400MHz bus, 1MB L2 cache) should be a significant upgrade over the G4 processor in the iBook and Mac mini. It's also cool enough for the small form factors and cheap enough for this price range.
-
Re:my take on the new PowerMacsThe nVidia 6600 is more comparable to an ATI X600, both of which are a generation beyond the ATI 9800 (and the nVidia 5900).
Actually, the Radeon X600's architecture is from the same generation as the Radeon 9600. The X600 is the PCIe version of the 9600. The generation beyond the 9600/X600 architecture is the X700.
I freakin' hate these inconsistent naming schemes. The Radeon 9200's architecture is the same as the Radeon 8500 (DirectX 8 for Windows users). The NVIDIA GeForce 4 MX uses the GeForce 2 architecure (no pixel/vertex shaders).
-
Re:my take on the new PowerMacsThe nVidia 6600 is more comparable to an ATI X600, both of which are a generation beyond the ATI 9800 (and the nVidia 5900).
Actually, the Radeon X600's architecture is from the same generation as the Radeon 9600. The X600 is the PCIe version of the 9600. The generation beyond the 9600/X600 architecture is the X700.
I freakin' hate these inconsistent naming schemes. The Radeon 9200's architecture is the same as the Radeon 8500 (DirectX 8 for Windows users). The NVIDIA GeForce 4 MX uses the GeForce 2 architecure (no pixel/vertex shaders).
-
Re:Radeon 9550 vs. 9200I tend to think of the Radeon 9200 as comparable to the GeForce FX 5200,
The Radeon 9200 (OpenGL 1.3, DirectX 8.1) is actually an updated Radeon 8500 and is comparable to the GeForce 3. ATI just re-used their previous generation's high end technology into their low end product. The GeForce FX 5200 (OpenGL 1.4, DirectX 9) is comparable (in features) to the Radeon 9550, but slower.
although the latter does support Core Image.
Core Image seems to require an OpenGL 1.4 GPU, which is probably why it requires a GeForce FX 5200 or Radeon 9600 minimum. Even though the 9550 is not listed among the supported GPUs, I think it should work since it's just a slower version of the 9600.
-
Re:I know the right name:I know the right name: Windows 64.
After all, this new version of Windows is likely going to be designed mostly to support the x86-64 CPU instruction set that AMD pioneered and Intel has begun to use, so this sets a break with the 32-bit past of previous Windows versions.
Interesting, but I think it will still be way too early (H2 2006) to make "a break" with the 32-bit past, present, and future. Remember, an AMD64 or EM64T processor is a minimum system requirement for x64 versions of Windows. The vast majority of the "mainstream" desktops sold today still use 32-bit processors (Celeron, Sempron, P4 5xx). The next Pentium M-like core (Yonah, Q1 2006) will still be 32-bit, and this architecture won't add EM64T until Merom (Q4 2006).
When Vista is released, most (maybe all) new Intel notebooks will still have 32-bit processors. Also, mainstream support for Windows XP Home (no extended support) ends December 31, 2006 (I'd be surprised if they didn't extend it), so there will be millions of potential 32-bit upgraders that Microsoft can't ignore. There are also all those businesses that still haven't been convinced to upgrade from Windows 2000 and have modern 32-bit hardware.
-
Re:I know the right name:I know the right name: Windows 64.
After all, this new version of Windows is likely going to be designed mostly to support the x86-64 CPU instruction set that AMD pioneered and Intel has begun to use, so this sets a break with the 32-bit past of previous Windows versions.
Interesting, but I think it will still be way too early (H2 2006) to make "a break" with the 32-bit past, present, and future. Remember, an AMD64 or EM64T processor is a minimum system requirement for x64 versions of Windows. The vast majority of the "mainstream" desktops sold today still use 32-bit processors (Celeron, Sempron, P4 5xx). The next Pentium M-like core (Yonah, Q1 2006) will still be 32-bit, and this architecture won't add EM64T until Merom (Q4 2006).
When Vista is released, most (maybe all) new Intel notebooks will still have 32-bit processors. Also, mainstream support for Windows XP Home (no extended support) ends December 31, 2006 (I'd be surprised if they didn't extend it), so there will be millions of potential 32-bit upgraders that Microsoft can't ignore. There are also all those businesses that still haven't been convinced to upgrade from Windows 2000 and have modern 32-bit hardware.
-
Re:I've said it once...The funny thing is that AMD is planning on releasing Quad Core chips in Q1 2006. Intel's DC Dothan may be too little, too late.
Do you really think AMD and chipset makers can deliver 65nm quad core CPUs to the desktop platform (Athlon, not Opteron) in the same time frame that Intel can deliver a desktop version of Yonah? Intel has been showing prototypes of Yonah since February and will supposedly ship Yonah to notebook manufacturers in Q4 2005. AMD hasn't shown or officially announced the availability of their quad core CPU, except for some hints to an Inquirer reporter at a trade show.
-
Re:This obviously means no Powerbook G5s- A switch to Intel CPUs. That likely means Pentium-M or Celeron-M in their small-form-factor (Mini, iMac, eMac) and notebook (iBook, PowerBook) computers, and potentially Pentium-4 in their desktop line.
Don't forget that this transition will likely be taking place from mid-2006 (low end) to mid-2007 (high end). From the article:
Apple plans to move lower-end computers such as the Mac Mini to Intel chips in mid-2006 and higher-end models such as the Power Mac in mid-2007, sources said.
That likely means (for the high end) some next-generation Intel CPUs like Merom (notebooks) and Conroe (desktops). On the other hand, the PowerBook seems to be the Mac most in need of a modern CPU, so I can't see Apple waiting for Merom (due H1 2006) or a low-heat G5. If these crazy "Intel inside Mac" rumors are true, I think Apple would want Yonah (dual-core 65nm Pentium M) inside the PowerBook. -
Re:This obviously means no Powerbook G5s- A switch to Intel CPUs. That likely means Pentium-M or Celeron-M in their small-form-factor (Mini, iMac, eMac) and notebook (iBook, PowerBook) computers, and potentially Pentium-4 in their desktop line.
Don't forget that this transition will likely be taking place from mid-2006 (low end) to mid-2007 (high end). From the article:
Apple plans to move lower-end computers such as the Mac Mini to Intel chips in mid-2006 and higher-end models such as the Power Mac in mid-2007, sources said.
That likely means (for the high end) some next-generation Intel CPUs like Merom (notebooks) and Conroe (desktops). On the other hand, the PowerBook seems to be the Mac most in need of a modern CPU, so I can't see Apple waiting for Merom (due H1 2006) or a low-heat G5. If these crazy "Intel inside Mac" rumors are true, I think Apple would want Yonah (dual-core 65nm Pentium M) inside the PowerBook. -
Re:This obviously means no Powerbook G5s- A switch to Intel CPUs. That likely means Pentium-M or Celeron-M in their small-form-factor (Mini, iMac, eMac) and notebook (iBook, PowerBook) computers, and potentially Pentium-4 in their desktop line.
Don't forget that this transition will likely be taking place from mid-2006 (low end) to mid-2007 (high end). From the article:
Apple plans to move lower-end computers such as the Mac Mini to Intel chips in mid-2006 and higher-end models such as the Power Mac in mid-2007, sources said.
That likely means (for the high end) some next-generation Intel CPUs like Merom (notebooks) and Conroe (desktops). On the other hand, the PowerBook seems to be the Mac most in need of a modern CPU, so I can't see Apple waiting for Merom (due H1 2006) or a low-heat G5. If these crazy "Intel inside Mac" rumors are true, I think Apple would want Yonah (dual-core 65nm Pentium M) inside the PowerBook. -
Sonoma's successor is Napa, then Santa RosaSonoma = California Wine Country
The Sonoma platform will be followed by the Napa platform. Napa will be followed by the Santa Rosa platform (Santa Rosa is a city in Sonoma County, not a wine region). According to Tom's Hardware, the platform preceding Sonoma had the code name "Carmel."
I bet this is all very uninteresting and off-topic. Go ahead and mod me down if appropriate.
-
Sonoma's successor is Napa, then Santa RosaSonoma = California Wine Country
The Sonoma platform will be followed by the Napa platform. Napa will be followed by the Santa Rosa platform (Santa Rosa is a city in Sonoma County, not a wine region). According to Tom's Hardware, the platform preceding Sonoma had the code name "Carmel."
I bet this is all very uninteresting and off-topic. Go ahead and mod me down if appropriate.
-
Sonoma's successor is Napa, then Santa RosaSonoma = California Wine Country
The Sonoma platform will be followed by the Napa platform. Napa will be followed by the Santa Rosa platform (Santa Rosa is a city in Sonoma County, not a wine region). According to Tom's Hardware, the platform preceding Sonoma had the code name "Carmel."
I bet this is all very uninteresting and off-topic. Go ahead and mod me down if appropriate.
-
Re:AMD whupped Celeron long sinceAMD doesn't need to release a whole new line of processors just to compete with the Celeron -- they've had the Celeron beaten for years.
Of course, the submitter's reference to the Celeron was a joke (Turion = Asparagus, Celeron = Celery). I mostly agree that AMD has kicked Intel's arse in budget CPUs in the past. But I don't think AMD currently beats Intel in every budget segment.
Consider Anandtech's conclusion from a Sempron vs. Celeron test they did last July
Also consider current pricing (from Newegg) for the Sempron and the Celeron D. In Anandtech's benchmark results, the Socket 754-based Sempron 3100+ did beat the Socket 478-based Celeron D 335 in most of the bechmarks that count. However, the Sempron 3100+ costs $123/$108 (retail/OEM) while the Celeron D 335 costs $109/$89. The benchmarks also showed that the Socket A-based Sempron 2800+ ($109/$99) was about even with the Celeron D 335, but would you choose an aging Socket A platform (PCI, AGP, IDE, 333MHz FSB) over a modern platform (PCI Express, SATA, 533/800MHz FSB) that you can get with the Celeron D?
Of course, we're talking about building our own desktops, which is very different from what the big-name computer makers offer. Us home builders would choose nForce or 915 chipsets for Sempron and Celeron D CPUs. HP and Dell are more likely to offer VIA/SiS/ALi chipsets for Semprons and 865/845 chipsets for Celeron D. Ugh.
If I was building a budget desktop, I would choose a Socket 775-based Celeron D over a Socket A or Socket 754-based Sempron. I value the whole platform just as much (if not more) than the CPU itself. If AMD made a Socket 939-based Sempron, I'd reconsider.
I'll be interested to see how this unfortunately named "Turion" chip compares to the PentiumM.
Back to the article's topic (notebook CPUs), it looks like AMD will not have an answer to Intel's Celeron M. The Celeron M is based on the Pentium M core and performs almost equivalently clock-for-clock in Tom's Hardware benchmarks. Also note that Intel's Sonoma platform (533MHz bus, PCI Express, DDR2, GMA900 graphics, HD Audio) is about to be lauched.
-
Re:Transistors
NV40 GPU: 222 M transistors
boyah! -
Re:6600 or 6800LE?
What's weird is that nVidia already _does_ have a $200 variant of Geforce6 - the Geforce 6800LE. It's essentially a lower-clocked (GPU and RAM) 6800 with only 8 pipes (so, half of what the 6800GT/U has).
Still, it'll be nice to see nVidia actually try to deliver a better price/performance ratio than ATI for once.
I hope this forces ATI to respond with a "mainstream" retail GPU (~$200) based on some form of the Radeon X800 core. Currently, the cheapest retail GPU based on the X800 core (X800 Pro) is around $400. The cheapest OEM GPU is the X800 SE (~$300), which OEM-only.ATI's current mainstream GPU, the Radeon X600, is based on the old Radeon 9600 core. ATI's only announced update for mainstream GPUs is RV410, which is just a 110nm version of X600. This will not do. ATI needs a slightly lower-clocked version of the X800 SE to compete with NVIDIA's $200 GeForce 6 offerings.
Compared to ATI's previous "SE" GPU's, the X800 SE seems pretty nice on paper. Unlike previous SE's, the memory bus isn't chopped in half. For those that haven't read about it:
- X800 SE: 8 pixel pipelines, 256-bit memory bus
- X800 Pro: 12 pixel pipelines, 256-bit memory bus
- X800 XT: 16 pixel pipelines, 256-bit memory bus
-
Re:6600 or 6800LE?
What's weird is that nVidia already _does_ have a $200 variant of Geforce6 - the Geforce 6800LE. It's essentially a lower-clocked (GPU and RAM) 6800 with only 8 pipes (so, half of what the 6800GT/U has).
Still, it'll be nice to see nVidia actually try to deliver a better price/performance ratio than ATI for once.
I hope this forces ATI to respond with a "mainstream" retail GPU (~$200) based on some form of the Radeon X800 core. Currently, the cheapest retail GPU based on the X800 core (X800 Pro) is around $400. The cheapest OEM GPU is the X800 SE (~$300), which OEM-only.ATI's current mainstream GPU, the Radeon X600, is based on the old Radeon 9600 core. ATI's only announced update for mainstream GPUs is RV410, which is just a 110nm version of X600. This will not do. ATI needs a slightly lower-clocked version of the X800 SE to compete with NVIDIA's $200 GeForce 6 offerings.
Compared to ATI's previous "SE" GPU's, the X800 SE seems pretty nice on paper. Unlike previous SE's, the memory bus isn't chopped in half. For those that haven't read about it:
- X800 SE: 8 pixel pipelines, 256-bit memory bus
- X800 Pro: 12 pixel pipelines, 256-bit memory bus
- X800 XT: 16 pixel pipelines, 256-bit memory bus
-
Re:6600 or 6800LE?
What's weird is that nVidia already _does_ have a $200 variant of Geforce6 - the Geforce 6800LE. It's essentially a lower-clocked (GPU and RAM) 6800 with only 8 pipes (so, half of what the 6800GT/U has).
Still, it'll be nice to see nVidia actually try to deliver a better price/performance ratio than ATI for once.
I hope this forces ATI to respond with a "mainstream" retail GPU (~$200) based on some form of the Radeon X800 core. Currently, the cheapest retail GPU based on the X800 core (X800 Pro) is around $400. The cheapest OEM GPU is the X800 SE (~$300), which OEM-only.ATI's current mainstream GPU, the Radeon X600, is based on the old Radeon 9600 core. ATI's only announced update for mainstream GPUs is RV410, which is just a 110nm version of X600. This will not do. ATI needs a slightly lower-clocked version of the X800 SE to compete with NVIDIA's $200 GeForce 6 offerings.
Compared to ATI's previous "SE" GPU's, the X800 SE seems pretty nice on paper. Unlike previous SE's, the memory bus isn't chopped in half. For those that haven't read about it:
- X800 SE: 8 pixel pipelines, 256-bit memory bus
- X800 Pro: 12 pixel pipelines, 256-bit memory bus
- X800 XT: 16 pixel pipelines, 256-bit memory bus
-
Wait until the R500 comes out Q4 2004...
http://endian.net/details.asp?tag=atir500
The R500 architecture will support DirectX 10, PS3.0, with 128-bit precision.
ATI has everything planned out. So it's a situation where nVidia has just come out with their next-generation card, but ATI has just sped up their last card to kill time until the super-high-end card comes out later this year. ATI has yet to reveal their true next-generation card, while nVidia's new generation card is already equalled by ATI's current one with simple feature increases. -
Re:Hundreds of thousands??
Pasted the wrong link the second time.
http://endian.net/details.asp?ItemNo=3906