|
|
|
|
20XX |
Q1 2006 01 Intel CPU Price list changes for newly introduced processors occurred on January 2nd. See the Intel CPU Prices page for additional information. Intel Core Duo (Yonah) CPU was released on January 5th. Yonah is part of the Napa platform and the successor to Dothan. It is built on a 65nm process and contains two CPU cores on a single die taking 151.6 million transistors (compared to around 140m in Dothan). Yonah features a 667Mhz FSB speed and a 2Mb L2 cache shared between the two cores. In addition to the featureset of Dothan, Yonah features an updated SSE instruction set (SSE3 support and SSE/SSE2 support tied to MicroOps Fusion, and throughput improvement in the SSE Decoder) along with improved performance of the x86 FPU. These improvements to the floating point performance of the processor are dubbed "Digital Media Boost" by Intel. Yonah also features improvements to power management and thermal output, meaning that the power consumption of Jonah will be similar to that of Dothan, despite the increased core size and complexity. Intel's Advanced Thermal Manager offers finer grained thermal management and Yonah will allow for one core to be slowed down independently of the other if power consumption or heat dissipation rises too much. Yonah also features Intel's Vanderpool technology and LaGrande Security technology, but does not contain x64 instruction set support in it's first incarnation. The initial members of Yonah are the T2300 (1.66Ghz), T2400 (1.83Ghz), T2500 (2.0Ghz), T2600 (2.16Ghz) and the Low Voltage L2300 (1.5Ghz) and L2400 (1.66Ghz). Initial benchmarking with Yonah versus an identically clocked Dothan show that the Dual Core Yonah is between 11% and 60% faster than Dothan whilst having equal or better battery life. Intel Core Solo (Yonah Single Core) was released on January 5th. The single core version of Yonah contains 2MB of L2 cache. The initial member of the Core Solo series is the T1300 (1.66GHz, 667MHz FSB). Intel Calistoga chipsets (955XM, 945PM & 945GM), part of the Napa platform for Yonah, was released on January 5th. Calistoga chipsets are support a 667Mhz FSB, DDR-2 667 and PCI Express. Calistoga will be paired with ICH7-M, featuring 4xSATA-300 ports, and the Golan (3945ABG) wireless chipset, supporting 80211a/b/g and WPA2. nVidia GeForce Go 7800 (G70M) was released on January 5th. The GeForce Go 7800 is based on the GeForce Go 7800 GTX, but runs at a lower clockspeed and features 20 pixel shader pipelines rather than 24. nVidia GeForce Go 7600 (G73M) was released on January 5th. The GeForce Go 7600 is based on the new G73 core, which will replace the GeForce 6600 series chips. The G73 core features 8 pixel processors, 5 vertex processors and 8 raster operation units (ROPs), and sports a 128-bit memory interface. The Go series can be clocked up to 450MHz core and up to 1GHz memory, although the exact frequencies are up to notebook suppliers. nVidia GeForce Go 7400 (G72M) was released on January 5th. The GeForce Go 7400 is based on the new G72, which will replace the GeForce 6200 series of chips. G72 is expected to feature 4 pixel shaders containing the enhancements introduced with the G70. ATI Mobility Radeon X1400 (M54) was released on January 5th. The Mobility X1400 is targetted at the Mainstream market. ATI Mobility Radeon X1300 (M52) was released on January 5th. The Mobility X1300 is based on the RV515 core powering the desktop Radeon X1300. Apple iMac Intel Edition was unveiled on January 9th, featuring Intel's Core Duo processor. AMD Athlon FX-60 Dual Core CPU was released on January 10th. Based around the Toledo core, the FX-60 is clocked at 2.6Ghz (compared to 2.4Ghz of the top end Athlon 64 X2 4800+) and contains 2x1MB L2 cache. This processor is the fastest planned for the Socket 939 platform. nVidia nForce4 SLI XE and
nForce4 Ultra MCP Intel Edition were released on January 17th.
The nForce4 SLI XE is designed for the Performance Mainstream segment
whilst the nForce4 Ultra MCP is aimed at the Mainstream market. Both
chipsets are based on the previous
nVidia nForce 4 Intel
Edition chipset, but have reduced featuresets. Whilst lacking ActiveArmour,
having 8 USB port support (compared with 10) and featuring lower
overclocking potential than their predecessors, both chipsets feature
Azalia HD Audio which was not present in previous chipsets. The Ultra
chipset also lacks SLI support, but for their target markets they are
feature rich, providing full support for a 1066Mhz FSB, RAID 0, 1 and 5,
Gigabit Ethernet and SATA II. nVidia GeForce 7300GS (G72) was
released on January 18th. The GeForce 7300GS is nVidia's first
desktop part to be built on a 90nm process and is a replacement for the
GeForce 6200 series. The GeForce 7300
GS features 4 pixel pipelines, 3 vertex pipelines and 2 ROPs, like the
GeForce 6200 series, but with an enhanced featureset and increased clock
speeds. The clock speeds of this part are 550Mhz core and 700Mhz memory
over a 64-bit bus. In addition to the featureset of the GeForce 6200
series, the 7300 features support for FP16 blending and the low-level
performance improvements introduced with the
G70 GPU. AMD CPU Price cuts occurred on January 23rd for Opteron processors. See the AMD CPU Prices page for more details. ATI Radeon X1900 (R580) GPU series was released
on
January 24th on a 90nm process. The R580 is a refresh version of
R520, featuring 384m transistors
compared to 321 for X1800, with the fundamental architecture only
slightly modified. The main differences between the R580 and R520 GPUs
are the number of pixel shaders, which has been tripled from 16 to 48,
and the slight reduction in ROPs (Raster Operation Units) which has
decreased from 16 to 12 for architectural reasons (4 pixel shaders,
consisting of 2 Vector ALUs, 2 Scaler ALUs and a Branch Execution Unit,
and - of course - one ROP per pixel shader engine). There has been no
change to the number of vertex shaders with R580, like the R520,
featuring 8 vertex shaders. Other improvements over the R520 include an
increased hierarchical Z cache (up 50%), which is primarily to help keep
performance high at extreme resolutions (above 1600x1200), and
improvements to the texture fetch units which now support a feature
called "Fetch4", first introduced with the X1600 and X1300 GPUs, which
allows a single texture address unit to read four texture locations at a
time for single channel data. This technique can speed up various soft
shadowing algorithms. Radeon X1900XT - 625Mhz core, 512MB of 1.25ns GDDR3 memory
clocked at 1450Mhz . $549. Initial benchmarks show that the X1900XTX is undoubtedly the fastest graphics card to date, offering up to an 80% performance improvement over the previous champion, the GeForce 7800GTX 512MB. Generally speaking, however, the performance is around 10-30% better at extreme resolutions with the highest image quality. Intel 945GT chipset for Yonah was released on January 30th. The i945GT is functionally identical to the i845G, but provides an interface to Yonah rather than the P4. This chipset is expected to be used on forthcoming Mobile On Desktop motherboards. Internet Explorer 7 Beta 2 was released on January 31st. This release is available for public download. See the IE7 Roadmap entry for more details. 02 nVidia GeForce 7800GS was released on February 2nd for the AGP platform. Based around the G70 core of the GeForce 7800GTX, the GS features a reduced 16 pixel pipelines and 6 vertex pipeline. The GeForce 7800GS is clocked at 375Mhz core interfacing to 256MB, 1.2Ghz GDDR3 memory over a 256-bit interface. AMD CPU Price cuts occurred on February 13th for Athlon 64 X2 processors. See the AMD CPU Prices page for more details. Windows Defender Beta 2 was released on February 14th. This AntiSpyware product is the successor to Microsoft AntiSpyware beta 2, based on the GIANT AntiSpyware codebase. 03 ATI Crossfire Xpress 3200 (RD580) north bridge was released on March 1st. This chipset is ATI's answer to nForce SLI-32 solutions, providing 44 PCI Express lanes from the North bridge alone (nVidia uses two chips to achieve the same solution, with SLI using 16 from the 'North' bridge and 16 from the 'South' bridge). An additional feature of RD480 is that the HTT speed can be increased to significantly more than 1Ghz - a first for an AMD chipset. RD580 can be combined with any RX480 series south bridges - initially, with ATI SBs, SB450 but later in Q2 with SB600. AMD Opteron 185, 285 and 885 were released on March 6th, with the 285 and 885 having immediate availability and the 185 becoming available in 30 days. These CPUs run at a clock speed of 2.6Ghz and feature 2x1MB L2 cache. ATI Radeon Mobility X1800 / X1800XT (M58) was released on March 7th. The mobility model is based on the Radeon X1800 XT GPU, optimised for power usage. The two members of the X1800 series are: Radeon Mobility X1800 - 450Mhz core, 1Ghz memory. nVidia GeForce 7900 series (G71) GPUs are expected to be released on March 9th and will be produced on a 90nm process (compared with 110nm for the G70). The G71 core features 278 million transistors, compared to 302 million for the G70 - the difference being due to the re-architecture of the GPU for 90nm and the reduction in pipeline depth as the 90nm process can hit target frequencies more easily than the 110nm process. The successor to the GeForce 7800 series, the G71 features enhanced clock speeds over the 7800 series, but offers the same 24 pixel shader, 8 vertex shader, 16 ROP configuration and is functionally identical. Whilst not as innovative as ATI's X1900 architecture, the G71 series does offer virtually identical current and near-future game performance to this series. Games in the mid term and long term future, however, are likely to run faster on ATI's platform. It seems that nVidia's goal is to offer current game performance at a lower price point to ATI. This can be done as the G71 has a die-size of 196mm2 compared to 352mm2 in the R580 (278M transistors compared with 384 million). There are two versions of the GeForce 7900 available, differing only in their clock speeds: GeForce 7900 GTX 512MB - 650Mhz
core, 512MB 1.6Ghz GDDR3 memory. $500+. Initial benchmarks suggest that the 7900 GT offers slightly higher performance than the 7800GTX. This is unsurprising as the specifications are very similar (identical shader and ROP counts, but with the 7800GTX offering slightly lower clocks of 430Mhz core, 1.2Ghz memory). The 7900 GTX, however, offers performance very similar to ATI's Radeon X1900XTX and outperforms it in a dual card setup (i.e. SLI vs Crossfire). nVidia GeForce 7600 (G73) GPU was released on March 9th . This GPU features 12 pixel shaders, 5 vertex shaders and 8 ROPs, and is based around the 90nm G71 core. The memory interface is 128-bit wide - half that of the 7800/7900 series but the same as the 6600 series. There are expected to be two versions of the GeForce 7600 series - the GT and the GS - with the latter available on March 22nd: GeForce 7600 GT - 12 pixel pipelines, 5 vertex shaders. 560Mhz core, 1.4Ghz 128-bit memory (22.4GB/s). $180 - $230. The 7600GT's closest competition is the X1600XT (which it destroys in benchmarks) and the newly released X1800 GTO. Generally speaking the 7600GT and X1800 GTO offer similar performance. With AA/AF enabled the X1800 GTO tends to slightly outperform the 7600GT. However, with no AA/AF the 7600GT almost always outperforms the X1800 GTO by a fair margin. This is not surprising if you look at the core clock speeds vs the memory bandwidth of these cards. In nVidia's favour the 7600GT does not require an external power connector and it is priced around $50 cheaper, making the GT an affordable and fast mid range part. ATI Radeon X1800 GTO was announced on March 9th with availability starting on March 31st. The Radeon X1800 GTO is based on the R520 core, but has a pixel shader quad (4 pixel shaders & 4 ROPs) disabled for a configuration of 12 pixel shaders, 12 ROPs and 12 texture units with 8 vertex shaders. The core interfaces to 256MB of 1GHz GDDR3 memory, providing 32GB/s of memory bandwidth. The clock speeds of the GTO is identical to that of the X1800XL - 500Mhz core and 1Ghz memory - and the GTO is, in fact, the same product with a disabled pixel shader quad. Initial benchmarks suggest that the X1800 GTO is slightly faster, on average, when compared to nVidia's GeForce 7600GT. The X1800 GTO is expected to retail for $249 - around $50 higher than nVidia's 7600GT. Intel Dual Core Xeon LV (Sossaman) CPU was released on March 14th. Sossaman is the Xeon processor based around a 65nm Dual Core Yonah core. Sossaman is a low power processor, initially available in 2Ghz and 1.66Ghz revisions. The 2Ghz revision has a TDP of 31W, with the 1.66Ghz version having a TDP of just 15W. nVidia GeForce 7600 GS was released on March 22nd. The 7600GS is based on the G73 core but runs at slower clock speeds with a 400Mhz core clock and an 800Mhz memory clock over a 128-bit interface. Initial benchmarking shows that the 7600GS will outperform a GeForce 6600GT by around 10-15%, with the GeForce 7600GS offering performance around half way between a GeForce 6800GS and a GeForce 7800GT without AA/AF (the lower memory bandwidth will certainly hinder performance with AA / AF enabled, making it slower than the 6800GS), and generally better than the Radeon X1600 XT. The 7600GS is expected to retail around $129-$149. nVidia GeForce 7300 LE was released on March 22nd. The 7300LE is a reduced clock speed version of the GeForce 7300GS, featuring a core clock speed of 450Mhz and a memory clock of up to 666Mhz over a 64-bit bus, depending on the OEM. The 7600GS is expected to retail around $49-$69. Firefox 2.0 Alpha 1 was release on March 22nd. Intel Pentium Extreme Edition 965 (Presler) was announced on March 22nd, although it will not be shipping until April 30th. The 965 is based on the Extreme Edition Presler core and runs at a clock speed of 3.73Ghz. nVidia GeForce Go 7900GS was released on March 29th. Based on the G71 core, the 7900GS features 20 pixel processors and 7 vertex processors and has a core clock of 375MHz interfacing to 256MB of 1GHz GDDR3 memory over a 256-bit bus. Q1 Intel E8850 chipset for Paxville is expected to be released in Q1. The E8850 is part of the Truland platform, featuring support for Quad Channel DDR2 400 than can be arranged in a Mirror mode (like RAID1 to prevent data loss of a chip fails) and Hot Plug PCI Express. WinFS Community Technology Preview (CTP) is expected to be released in Q2. Visual Studio 2005 SP1 is expected to be released in Q1. Service Pack 1 is expected to contain purely bug fixes to VS 2005. |
|
|
|
|
|
Copyright © 1999-2006 Michael K. Warner. All rights reserved. No part of the content of this web-site may be reproduced in any form without prior written consent. Please send any comments or queries to mike@mikeshardware.co.uk. |