Intel puts five Xeons on death row

| Sunday, April 26, 2009

Intel's newly released Nehalem EP-core Xeon processors have inflicted their first casualties: five older Xeons are scheduled to be phased out.

According to Intel's Product Change Notification #109186-00 (PDF), you'll be able to buy these death-row Xeons in a retail box until January 9, 2010.
Click here to find out more!

The Xeons set for the chopping block all have 1333MHz frontside buses. The two parts with X in their product number are 45nm, 95-watt Yorkfield parts, the three without are 65nm, 65-watt Conroes.

The doomed processors are:

* Quad-core 2.66GHz Xeon X3350 with 12MB L2 cache
* Quad-core 2.5GHz Xeon X3320 with 6MB L2 cache
* Dual-core 3GHz Xeon 3085 with 4MB L2 cache
* Dual-core 2.66GHz Xeon 3075 with 4MB L2 cache
* Dual-core 2.33GHz Xeon 3065 with 4MB L2 cache

In microprocessor evolution as in nature, there are winners and there are losers. According to Intel, these parts are being phased out because "market demand ... has shifted to other Intel processors."


SOURCE:http://www.theregister.co.uk/2009/04/08/discontinued_xeons/

Intel's silicon brick roadmap

|

Intel gave us the lucky notches a launching of post-Nehalem giving of the instructions to the overall aspects of its short-term roadmaps of processor. Here 'fast excursion of SA by the synopsis of the matters.

Intel reassured us that it could see Moore 'progression of law of S being maintained for ten more approximately years with the contractings of process and other developments in its roadmaps. Inexorable rise in the transistor counts on the matrices, 731 million with quadruple-cores of Nehalem, should continue above this period. The great exit is people including/understanding how to employ parallelism.

Processors of waiter of Intel

Intel 'of the platforms of waiter of S groups covers several markets: high-output computer (the HPC), computer of Internet and the cloud, migration centres of RISC, storage and centre computer virtualized. Its constitutive blocks are central processing units of treatment, chipsets, the LAN, SSD/RAID, panels/KDKs and software. Its platforms of processor are Itanium (9000 series) for the waiters mission-critical, Xeon 7000s for the waiters at end high of company, Xeon 5000s for the traditional waiters of quadruple and duel-casing of company, and Xeon 3000s for the simple waiters of SME of casing. Obviously Xeon 7000s and Itanium compete with each one with the 'eyes of S but in Intel 's.

Much was done in Intel conceiving in devices support Virtual Server to obtain if all goes many apps functioning in virtual machines functioning as quickly as of the apps functioning in a non-virtualized waiter.

We had fast gambades by Xeon 5500 of the 'devices of S and then obtained with not very interesting, Westmere-EP, the processor of the second generation of Nehalem (5000 series) built with a forthcoming process 32nm. Westmere is basically a contracting of Nehalem on the process 32nm.

It will have six cores, improvements of hiding-place and better energy efficiency, and will be announced in 2010. Also it will be casing-compatible with the 5500, thus the suppliers of waiter can employ the councils of waiter and the existing processors of Westmere of catch directly inside. The graph showed a bus of NCV 2 Westmere binder with a transistorized disc of Intel X25-E, and a controller of Intel 10GbitE.

Intel has an analogy of tick-tock for its developments of technology. With drill is a contracting of process employing current microphone-architecture. The last drill was the movement with transformer technology 45nm current. With tock is the introduction of a new processor microphone-archictecture and Nehalem is the last event of tock.

It will be followed drill 32nm which will be in their turn followed by next microphone-architecture after Nehalem.

On before 7000 the 7400 is 6 a processor of the core 45nm which 's casing-compatible with the 7300 precedents. It will be followed of Nehalem-EX in second half of this year and then Westmere-EX a certain hour later.

Nehalem-EX, built on the process 45nm, is a processor of 8 cores designed for 4 platforms of casing, meaning 64 wire (2 wire/core always). It will employ an extensible interconnection of memory of Intel with shock absorbers and which Intel called the technology of the second generation of input-output with the virtualisation.

There will be technology of VT from Intel in the processor (VT-x), the tiddlywinks (VT-d) and for the network (VT-c). The idea is to obtain these components functioning better together so that the suppliers of waiter go all-Intel instead of employing third chipsets and interfaces of communications. Thus there 's no need for fibre Handle above adaptor Ethernet, a CAN, because these devices of initiating VT plus Intel 'of software FCOE of S will obtain the NIC from Intel 10GbitE squirting outside packages of FCOE at the speed close to wire to the devices direct-attaches.

There will be queues of device of virtual machine (VMDq) to improve the exit by unloading data matching with the NIC. In a waiter virtualized the virtual machine (VM) more or less of maintenance directly with the NIC without having to pass by the hypervisor. (Naturally this different VMs in the cycles from central processing unit of treatment of use of waiter means to carry out the work of FCOE however. For why the doesn of Intel 't goes FCOE-unload is beyond me. It tried to make the same thing with the TCP/IP unload, to say there was not no need, and for pretty lost much this argument.)

Intel presented the migration of cable thus of VMs can be easily transferred between various waiters from Xeon. Previously, notwithstanding the overall compatibility of instruction between 3000s, 5000s and 7000 Xeons, VMs could only be moved between the similar types of processor.

We can await a version of Nehalem of the 3300, probably the 3400 and probably at the end of this year or 2010.

Nvidia may not be licensed for future AMD chipsets

|

Nvidia does not benefit from a pleasant moment not in the businesses of tiddlywinks of mother chart at present. Not only Intel it went until continuing Nvidia after the company claimed that it had a licence to make chipsets for the core i7, but it maintaining glances as if Nvidia is not authorized to do with chipsets for future processors of AMD one or the other.

Betanews announces that vice-president Pat Patla of unit of company of the waiter of AMD claimed that �for 2010 advancing, the solutions leaving AMD will be AMD and on AMD currently.� The site indicates that Patla �did not believe that the situation of authorization between its company [AMD] and Nvidia would make it possible Nvidia to produce the chipsets which support future platforms of AMD.�

The comment was formulated with a press conference where AMD revealed its new six-core central processing units of treatment of �Istanbul� Opteron, as well as of future roadmaps of processor of Opteron which showed that plans released its 32nm �bulldozer� 16 central processing units of treatment of Opteron of core in 2011. According to betanews, Patla indicated that �we put 't intend to see new chipsets of Nvidia or Broadcom for achievements of waiter in 2010. But they will continue to support all the existing platforms advancing per 2010.

It is little clearly at present if Patla referred only to the platform of Opteron of AMD when it said that �the solutions leaving AMD will be AMD and on AMD�, or if this applies to the platforms of office of AMD as well. As well as to produce chipsets of SLI for AMD 'central processing units of treatment of office of S, Nvidia previously had a notable presence in platforms at end high of the work station of AMD too. The pro tiddlywinks 250 of the nForce 3 was made specifically for the central processing unit of treatment of Opteron of AMD, and Nvidia also developed the tiddlywinks of the nForce 680a SLI, which was the only tiddlywinks which supported the biprocesseuse platform of quadruple FX of AMD.

If it is filter indeed outside at the platform of office of AMD, then in Nvidia should have one difficult moment in the businesses of tiddlywinks. With the complaints of Intel that Nvidia does not have any licence to produce chipsets for central processing units of treatment of Intel with an integrated controller of memory, plus the plans of Intel to attach its central processing units of treatment of atom of �Pineview� to the chipsets of Intel with the integrated graphs, this would leave only Nvidia with the support for Nvidia old man and central processing units of treatment of AMD and Via the central processing unit of treatment of nano.

We asked AMD and Nvidia to comment on this, and to confirm if the restriction of licence applies just to the platform of Opteron of AMD, or with its platforms of office as well. We will give you an update as we have more information.

Intel Reveals New In-Store Concept Technologies at Retail’s Big Show

|

Tie profit of an high-execution and a piece of ideal efficiency that the company is already sold to the computer manufacturers, Intel Corporation today revealed a system with the detail of proof-of-concept of the sales outlet (position) which could lower the total costs of a store of the property while improving satisfaction of the customer.

Indicated to national convention with the detail of federation (NRF) in New York, the proof-of-concept of Intel shows more late in digital technology of signage and sales outlet in a factor of the shape of kiosk, and shows how technology can increase the experiment with the detail of purchases.

The obliging consumer's choices for the self-service filled of fuel by tendencies of e-business, the demonstration of kiosk includes the devices which match the advantages of the purchases of Internet, such as the access in suggestive real-time of sale of inventory, as well as to show promotions, details of product and reviews of customer. These devices, in their turn, facilitate the occasions of worms the high-sale and cross-country race-selling of associated sales, which will increase the result of a retailer and the fidelity of customer.

The demonstration of position is based on the mobile processor of very popular duet and ideal efficiency of 45nm Intel� Core.2 which is interior hundreds of million laptops and GCV of today. Compared with the platforms of position of legacy, the processor in the demonstration can allow more than one reduction of 70 percent of use required of power while always providing as much as the double the quantity of execution of treatment believed as former generations of processors.1 in oneself, of energy saving of this system of position are transmitted to the retailers by invoices of lower energy.

technologies INTEL-based will make sure that retailers well-are placed to enter by the current economy one succeeded and future advantageous, while providing to their customers more faster information and more easily than ever front,� said Joe Jensen, managing director, Intel included the Division of calculation. �We hope for the combination to improve an experiment with the detail of a customer, whereas the sales also of weapons joins more exact information and to reduce costs of operations, such as invoices of electricity and maintenance, is an irresistible proposal so that the retailers of the world benefit from what the innovation of technology of Intel must offer.�

The platform of proof-of-concept comprises the technology of vPro. of Intel� (vPro. of Intel�) with technology management of � of Intel activates and the lower costs of maintenance of offers, management improved of power and options of safety such as the diagnosis of system. The technology of vPro of Intel makes it possible administrators to control remote systems, making it possible to stop all the systems lasting during the night with energy saving of increase and the need of the technicians to go on the spot for the maintenance of system, having for result the decreased carbon emission and to lower operational expenditure for retailers.

The modular concept of the system of proof-of-concept thus supports easy levellings with the platforms of the second generation of central processing unit of treatment the retailers can remain in the synchro with last advances of technology. The demonstration of position is personnalisable with a panel of dismountable PC and the various modules which can be added above, removed, exchanged or replaced to meet businesses in evolution needs.

Intel Reveals New In-Store Concept Technologies at Retail’s Big Show

|

Tie profit of an high-execution and a piece of ideal efficiency that the company is already sold to the computer manufacturers, Intel Corporation today revealed a system with the detail of proof-of-concept of the sales outlet (position) which could lower the total costs of a store of the property while improving satisfaction of the customer.

Indicated to national convention with the detail of federation (NRF) in New York, the proof-of-concept of Intel shows more late in digital technology of signage and sales outlet in a factor of the shape of kiosk, and shows how technology can increase the experiment with the detail of purchases.

The obliging consumer's choices for the self-service filled of fuel by tendencies of e-business, the demonstration of kiosk includes the devices which match the advantages of the purchases of Internet, such as the access in suggestive real-time of sale of inventory, as well as to show promotions, details of product and reviews of customer. These devices, in their turn, facilitate the occasions of worms the high-sale and cross-country race-selling of associated sales, which will increase the result of a retailer and the fidelity of customer.

The demonstration of position is based on the mobile processor of very popular duet and ideal efficiency of 45nm Intel� Core.2 which is interior hundreds of million laptops and GCV of today. Compared with the platforms of position of legacy, the processor in the demonstration can allow more than one reduction of 70 percent of use required of power while always providing as much as the double the quantity of execution of treatment believed as former generations of processors.1 in oneself, of energy saving of this system of position are transmitted to the retailers by invoices of lower energy.

technologies INTEL-based will make sure that retailers well-are placed to enter by the current economy one succeeded and future advantageous, while providing to their customers more faster information and more easily than ever front,� said Joe Jensen, managing director, Intel included the Division of calculation. �We hope for the combination to improve an experiment with the detail of a customer, whereas the sales also of weapons joins more exact information and to reduce costs of operations, such as invoices of electricity and maintenance, is an irresistible proposal so that the retailers of the world benefit from what the innovation of technology of Intel must offer.�

The platform of proof-of-concept comprises the technology of vPro. of Intel� (vPro. of Intel�) with technology management of � of Intel activates and the lower costs of maintenance of offers, management improved of power and options of safety such as the diagnosis of system. The technology of vPro of Intel makes it possible administrators to control remote systems, making it possible to stop all the systems lasting during the night with energy saving of increase and the need of the technicians to go on the spot for the maintenance of system, having for result the decreased carbon emission and to lower operational expenditure for retailers.

The modular concept of the system of proof-of-concept thus supports easy levellings with the platforms of the second generation of central processing unit of treatment the retailers can remain in the synchro with last advances of technology. The demonstration of position is personnalisable with a panel of dismountable PC and the various modules which can be added above, removed, exchanged or replaced to meet businesses in evolution needs.

Intel launches mobile device chips, roadmap

|

With its forum of realizer of Beijing this week, Intel confirmed it is forwarding ultra mobile tiddlywinks for UMPCs (GCV ultra-mobiles) and MIDS (mobile devices of Internet). She also claimed to take a mobile processor of second generation based on silicon 45nm and High-K technology metal carries transistor.

Intel shared the first time its vision based on Windows of UMPC it there has one year, and indicated its SEMI vision Linux-based yesterday.

Intel 's new UMP2007 platform


Tiddlywinks of Intel UMP 2007
(Click to be increased)
Intel indicated the sound the platform ultra mobile 2007 tiddlywinks, formerly called McCaslin, will embark this year in devices of the dozen companies, including Aigo, Asus, Fujitsu, Haier, HTC, and Samsung. The tiddlywinks are based on Intel 'processors SA. 100 and A110 (formerly called Stealey ), has it purposely GU version of sound northbridge of laptop of tiddlywinks 945, and has U version of sound southbridge of laptop of hub of controller of input-output ICH7.


Intel 'mobile ultra diagram of platform 2007 of S
(Click to be increased)

The two parts of Stealey, doubled A100 and A110, are based on cores of Pentium M synchronized with 600MHz and 800MHz, respectively. Both have speeds of 400MHz FSB (bus on before-side), 512KB of the L2 hiding-place, and a package of 14 X of 19mm. Both are built on transformer technology 90nm and have a TDP (thermal power of design) of 3 Watts.

The northbridge 945GU supports until 1GB memory DDR2 with single channel, via a pair of interfaces of SO-DIMM. It has IGP of the GEN 3.5 Intel (processor of graphs integrated) synchronized with 133MHz, an interface of LVDS, and one 1 interface of PCI Express of lane.

Roadmaps

Going ahead, Intel takes has Silverthorne processor mobile device based on its industry-main thing process of silicon 45nm and High-K technologies metal carries transistor. Silverthorne will join to has with Poulsbo with piece of companion (northbridge/southbridge integrated), in Intel 'tiddlywinks mobile of second generation of S, called Menlow.

Interesting, the targets include the devices not only ultra-mobiles and mobiles, but also the desktop machines, work stations, and the waiters, illustrating the universal call of the power-effectiveness.

The additional prospect on Intel 'for the inking film news Beijing for S can be found in a perspicacious comment by the notable magazine Tim Bajarin of PC, here. Another report/ratio of store of PC on Intel 'of the inking film news of S today can be found here.

SiS Back On Stage: The SiS730 Chipset for Duron and Athlon

|

LOCATED to present the Top-League of the Set of chip-Manufacturers?

 LOCATED

During years the Taiwanese manufacturers of tiddlywinks weren 't able competing with much Intel 'products of tiddlywinks of S in terms of execution as well as success, but recently times changed. Via 'enormous success of S in 2000 gave Ali and LOCATED enough hope to return out of the darkness maintaining too. It is not astonishing that Intel will not threaten its business of central processing unit of treatment by providing chipsets for the processors competiting of Duron and Athlon. Continuous AMD to specify that he does not want to become a manufacturer of tiddlywinks, but a simply certain kind of indicator of technology. Instead of that, AMD wants mainly to show to the manufacturers tiddlywinks which tiddlywinks of Athlon should resemble.

Via was the first company to ensure proactivement of the chipsets for processors of AMD, this fact being one of the sweeping forces to establish the two processors of AMD on the market and themselves as a one of the higher suppliers of tiddlywinks today. Now, more than one year after the moment when Athlon 'successes of S were still dubious, Ali and LOCATED enter the market of AMD-punt-form finally too. The laboratories of Acer jumped the first two generations of Athlon and provides to its MaGiK1 tiddlywinks the support for GDR SDRAM and the recently announced processors of Athlon with 133 megahertz FSB. Via also works with its chipsets of GDR for Athlon (Apollo KT266). The first tiddlywinks of the integrated systems by silicon (LOCATED) are called 730s and can be called a chief of work of integration. A simple piece of BGA (672 balls) accomodates all the functions for which the other companies need at least two pieces. This interesting product is most certainly in value a glance.

Devices of the tiddlywinks of LOCATED 730S

 Devices

The list of device is indeed completely long. It is particularly interesting because the 730 is composed of only one simple piece of BGA. The LOCATED one decided to remain with memory of SDRAM, which gained 't is any true disadvantage for the next months. Please also read Tom 'evaluation of S of the tiddlywinks of KT133A. The reference marks there prove rather clearly that only the higher speed of FSB provides an increase in execution already. The clocks of bus of processor of 100 and 133 megahertz (this last for new C-Athlons) are supported by SiS730, giving you the option to use all the central processing units of treatment available of Duron and Athlon from 500 to 1200 megahertz and over there.

Controller of host centre

  • 1.8 V core with 3.3 V mixed and 5 V CMOS
  • Support of clock of bus of processor of 100 or 133 megahertz
  • Support of AMD PowerNow!
  • Support of synchronous or asynchronous buses of processeur/d' clock of memory

Interface DRACHMA

Port of AGP

Devices of the tiddlywinks of LOCATED 730S, continuous

Interface NCV

Controller of Busmaster ide

Accelerator of graphs 2D/3D integrated

Acoustics of NCV

Synthetizer of Wavetable

Management auditor of power

Other devices

Fast Ethernet

Functional diagram

 Block

The functional diagrams look at usually enough the more or less even thing. He 's simply interesting to see that all was integrated in a piece.

Also for casing 370: SiS630 and SiS635

 Moreover

The LOCATED one also offers the 630, the tiddlywinks 630S and 630E during a few months now, which includes the same basic devices. The principal difference is their support of the processors of Pentium III and Celeron.

If you want more information on logic of core of LOCATED, please throw a glance in the following page on the Web site of LOCATED:

Systems Corp. - product, logic of Silicon Integrated of core

The LOCATED one also works on tiddlywinks of GDR for systems of Pentium III and Celeron. With a little of chance, we could also see tiddlywinks of Athlon with GDR supporting as well.

Solution or complete platform 3D?

To a certain extent the chipsets of LOCATED are the products of Bohemian. Impressive being integrated, they are much less expensive to produce than solutions of two-piece. Consequently they seem predestinated for the inexpensive solutions of office, because they come with all the components which a commercial computer could have need: The graphs standard 2D/3D functions, a fast interface of ide, noise and even a controller of network - all is included for free. To have all this on the mother chart clearly helps to save the costs for the complete machine.

Astonishing enough by LOCATED decided to include a port of external AGP as well. Thanks to that, you can equip a mother chart SiS730 with a fast graphics card 3D, manufacturing a machine of play from it. I can imagine that many users could be interested by a cheap mother chart with these tiddlywinks instead of expensive buying one Via the solution. In this moment it is clearly more convenient than the gamers to save up to $50 on the mother chart and invest this money in a faster graphics card. The total execution will be usually better outward journey in this way.

By studying the reference marks satisfy take into account which I employed completely a slow processor, Duron 700 megahertz. That 's one of the cheaper models available ones in this moment. Nevertheless this central processing unit of treatment is rather quickly to produce good rates of reinforcement with the plays 3D currents - provided that you employ a good panel of the three-dimensional charts. The applications of Windows function usually rather quickly in any event, so that the software as MS Office can be left out of these considerations.

Gift 't forget that these thoughts are based about the fact that SiS730 is a solution of budget. He 's not tiddlywinks for the systems at end high.

Stability and drivers

I remember that there are a few years much people complained about the exits about reliability and stability of the platforms based above Via, LOCATED, VLSI, UMC and chipsets of Ali. While the VLSI was bought by Philips and UMC does not conceive any more of the chipsets, the others three recently provided the reliable and stable products for a completely certain hour now. I remember mother charts prototype to run so unstable that you had to spend all the night to obtain finished reference marks. The council of the SiS730 reference caused simple hang up again or no other trouble of the whole, which was a pleasant surprise, considering that it is always a panel of prototype. I must however complain about LOCATED the 'support about driver. Naturally they have their own AGP and visual drivers, but I missed a driver of refining of ide, which could have improved the execution of hard drive.

Does Intel have reasons with the threatened feeling?

Sometimes the domestic users do not realize owing to the fact that a great quota of all the sales of computer is employed by companies. Only one third of all the GCV can be found in the houses. If it were your work of order several dozen or even hundreds of computers for your company, you would start to make pointed calculations of the prices of razor.

Up to now there weren 't much of chances to assemble a bottom-of-the-range computer of Duron as cheap as a system based on Intel 's Celeron. The reason is very simple. There wasn 't any inexpensive tiddlywinks with the integrated graphs available. This situation looks at if you want to accumulate INTEL-based �value � - the completely different computer. The decision makers could select Intel 810 or 815 the tiddlywinks, Via 's ProSavage PM133 or Ali Aladdin TNT2. All include at least the functions of graphs.

If you want to build one on inexpensive computer on a processor of Duron, you recently were likely to employ Via 's ProSavage KM133, which also comes with the integrated graphs. Unfortunately, ProSavage KM133 only became available recently, and before the purchasers usually permuted with a solution INTEL-based.

SiS730 seems to be the perfect inexpensive solution. Our tests were very painless, proving enough well that the tiddlywinks seem to be matured. There are many reasons for which I would prefer certainly a system of Duron to a system of Celeron. The processor always could offer a much better report/ratio of price/performance. Celeron is limited to the clock of bus of processor of 66 megahertz (to 800 megahertz him 's 100 megahertz finally), whereas Duron functions to 100 megahertz double-pumped. Finally Duron can be associated with memory with PC100 or PC133 SDRAM, whereas the systems of Celeron usually function with memory of 66 megahertz. Moreover, there is from now on no difference of the prices between 66 megahertz and 133 memory of megahertz SDRAM. The question about the speed of memory is very important forchipsets with the integrated graphs, because the execution of graphs depends largely on the bandwidth of data between the core of graphs and of main memory, which is partly used as memory of graphs.

Testing facility

Material
Central processing unit of treatment AMD Duron 700 megahertz
RAM 1x 128 Moctet SDRAM (corsair)
PC133, 7ns, CL2
Hard drive IBM DeskStar 75GXP,
DTLA307030, 30 GIGAOCTETS
7200 t/mn
Graphics card nVIDIA GeForce 2 pro
Memory of 64 mbs GDR, 5 NS
Drivers and software
Drivers of HDD/Motherboard/AGP Via 4.26 complete
VIA AGP 4.04
LOCATED AGP 1.02
Graphics card Drivers 6.47 of reference of NVIDIA
Drivers 5.12.01.102 0 of graphs of LOCATED
Version of DirectX 8.0
Operating system Of Windows 98, 4.10.2222 has
Windows 2000 pro, 5.10.2195
SP1
Reference mark and arrangements
Arena of tremor III Version with the detail
line of order = s_initsound 0 of the cd_nocd 1 +set of +set
The graphs detail overall �normal �, 640x480x16
Using of reference mark �Q3DEMO1 �
MDK2 Downloadable version of demonstration
line of order = - timedemo
640x480x16
Sysmark 2000 Correction 4B
1024x768x16
ViewPerf Version 4.1.2
1280x1024x16, 75 hertz
Refresh rate 85Hz

Reference mark

Locate bandwidth of memory

 Memory

I had the practice SiSoft Sandra 6.0 to obtain some information of execution on the interface of memory. Everyone will agree with me that come does not seem to be a good interpreter. Control left 'of S the real-world evaluates now to see how this affects the daily software 'of S.

BAPCo SYSmark 2000

 BAPCo

In my opinion BAPCo 's SYSmark is always the best continuation of evaluation to determine the total execution of businesses. However, the real numbers are rather theoretical, because even a machine of 300 megahertz can provide execution enough to treat documents, diffusion covers or similar substance. The Green line is the result of reference mark using the integrated engine of graphs of SiS730. Blue represents the result using a pro graphics card of GeForce 2. As envisaged, the advantage of a so expensive graphics card above the integrated engine is not interesting the mention.

BAPCo SYSmark 2000 - Windows 98

 BAPCo

Because you could see in the front diagram, the new handset of SiS730 chips cannot beat Via 's KT133. That also applies to the new handset of chips of KT133A, because it has to only be updated to support 133 megahertz FSB (model Athlon C). Tiddlywinks of LOCATED 'of S 730 support already this bus of processor by defect.

The delta of execution between the LOCATED one and Via the tiddlywinks is between 3 and 12% out of Windows 98, according to the application. The intensive applications of memory like Bryce, elastic reality (modelling and returning 3D), the first (video publishing) or the paradox (data base software) suffer clearly from the bandwidth of LOCATED low SDRAM. However, you gained 't can obtain these points using a system of Celeron 700.

BAPCo SYSmark 2000 - Windows 2000

 BAPCo

The situation is completely the same one in Windows 2000. Just the result of point of power is considerable higher using has Via the mother chart KT133. You can also see that both chipsets are almost on the par in Photoshop 5.5, aspiration 9 of Corel, first 5.1 and word 2000.

Demonstration MDK2

 MDK2

MDK2 is a gunner of action of Direct3D-based. We employed the version of demonstration for the evaluation: As you can see, the engine of graphs integrated of the tiddlywinks of LOCATED is not exactly a large interpreter. By employing the graphics card of GeForce 2, there is very small difference between SiS730S and Via 's Apollo KT133.

Packing of truck of benz of Mercedes

 Mercedes

The truck packing the demonstration offers the very good quality of image if you light all the possible devices 3D. However, that would have made the play and the reference mark unplayable on the engine of graphs of LOCATED, which 's why I carried out these tests with arrangements of defect. Even with 640x480, this play is not recreation exploiting our configuration of system of reference mark. 22 images a second are not sufficient for the play of play. After connection in the pro chart of GeForce 2, you can easily play MBTR with until 1280x1024 on the LOCATED one and Via the system. With 1024x768 and 1280x1024, this play functions even more quickly on the platform of LOCATED! As soon as you increase the resolution, you can see how the bandwidth of lower memory of SiS730 has gradually more impact on the rate of reinforcement.

Arena of tremor III

 Tremor

Tremor III is absolutely unplayable with the engine of graphs of LOCATED. Even with 640x480, the system could not even mark 20 fps. Still, using the nVIDIA fast GeForce 2 pro thorough of council to the top of the rate of reinforcement 5-10 times. Via 's KT133 is approximately 10% faster than SiS730.

SPECviewperf 6.1.2

 SPECviewperf

The differences in execution between the LOCATED one and Via the tiddlywinks are much larger using the high reference mark at end SPECviewperf d' OpenGL. It can also be employed like test for the stability of system, because it will most probably break if a component of system surely does not function.

The design engineering (DRV-07) more quickly runs approximately 35% on Via the council than on SiS730 - a clear homage to the better execution of memory of tiddlywinks KT133. All other applications function certainly more quickly on panel KT133 as well. If you envisage to launch such programs at end raised, you should spend the higher price of has Via the mother chart oftiddlywinks KT133.

Conclusion

The revision of the mother charts of prototype formerly was completely a piece of work. Not this one. No hang rises, aucunes resulting from driver. All was right as with an entirely matured product. I guess that the LOCATED one spend much time to optimize and refine the tiddlywinks, because there was not apparently no need to precipitate it at the market as quickly.

Is the result with the Juste of what the mediums of the businesses had need: Relatively fast, reliable and well-comprised tiddlywinks. No other manufacturer offers a product with the same range of device as the family SiS630 and SiS730.

Though the engineers of LOCATED count on the drivers of the defect ide of Windows, I 'm sour which specialized those can improve the total execution even more. There could also be a certain part to optimize the execution of memory as well. If you think of the comparisons of tiddlywinks of the whole the decade spent, the LOCATED one could be in a good situation. This time there is not any tiddlywinks competiting of Intel to exceed the product of LOCATED.

Segment finally the bottom of value/budget has now two chipsets different from Athlon/Duron to select, Via 's ProSavage KM133 and SiS730. I expect that the LOCATED one cheaper although better is described. With these two chipsets available, the sector of calculation of businesses is not any more according to chipsets or processors of Intel. There is much more behind him. The value of a commercial computer is increased by a processor of AMD Duron. You obtain a faster treatment unit, a faster memory and better tiddlywinks describe for the same price - isn 't which much?

The sector of house and execution-computer of office was the first of three sectors to enter by AMD, now AMD has good lucks write the segment of value. This freely leaves always only one part of the processors of AMD: Systems of work stations and waiter. That will change once AMD obtains the tiddlywinks of AMD760-MP ready to go.


ABit BL7 RAID vs IWill P4S

|

Please see more?
Of my last experiments which spanned the several generation of the councils, the first thing that I usually make is to seek the place of the port of FDD, the ports of ide and of connections of food of energy. With much groove panels 1, the proximity of the slits of DIMM and the size of the condensers can indeed block the use of larger a radiator and ventilator. Now, with parallel typically assembled of ventilators to the mother chart, this 's become an not-exit.


Overall picture of IWill P4S



My installation is composed of a full envelope of turn, with all my HDD assembled in 5 1/4 compartments. I noted that the placement of the ide puts in communication at its remote end, required the purchase of a longer cable of ide to connect itself upwards. The placement of the ports of ide close to the connection of food of energy would have been a better IMHO of position, although it can become rather with narrow there.


Others look at, of another angle



Nothing of nine here. He 's the same ones with all connections of casing 478.


Casing 478



IWill using a reward of bios of Phoenix in fact became part of technologies of Phoenix in September 1998. Products from both are now manufactured and jointly launched on the market.


The piece of BIOS



As always, the tiddlywinks i845 are cooled by a radiator. Note that the use of metal retains to hold the radiator in place.


The timid i845 being covered



Same Southbridge used on all the mother charts i845.


Southbridge which orders no matter what the doesn of Northbridge 't



With the difference of ABit BL-7 RAID, connections back of IWill P4S are completely the standard which you see on older designs of mother chart. The only reason that I can propose, is that ABit chose a design of council requiring reserved space particularly for a port of LAN in their councils. These manufacturing costs and of design in manner become reduced, so during a day a version of LAN is launched.


Connections back



CNR groove is removed. Not that I miss it in any event. If you precisely prove to be the rare one little using the slit of CNR, these Ain 't the council for you.


I do not see any slit of CNR



For those which improve in Pentium 4 or which emigrate more of AMD, you can have to obtain a new unit of food of energy. Pentium 4 employs a power of 300w ATX12V which comes with connectors from food from additional energy necessary for connections, as shown below.


Additional connections of power required


Chapter 42. I/O and The south bridge

|

We now looked at the various buses of input-output, and the bridge of the south of the tiddlywinks plays the central role in the transfer of the data on these buses. But, as mentioned front, the bridge of the south is helped by a small superb controller of input-output. Let briefly describe it to me.

Functions of the bridge of the south

The bridge of the south is a piece on the mother chart. If we want to look at one of the last models, we could take the bridge of the south of the tiddlywinks of Intel 915, which is designed for mother charts with processors of Pentium 4.

Diagram 191. Tiddlywinks 915G of Intel. The bridge of south ICH6 is towards the left.

A glance on the characteristics of the piece indicates that it contains the components and the functions following:

Component

Description

DMI

The direct interface of media is connection with the memory with a maximum bandwidth of 2 GB/sec.

PCI Express

bus Safety-speed for adaptors of input-output

Ports of NCV

Standard bus of input-output.

Serial ATA

Controller for to four hard drives of SATA

Storage of Matrix

The advanced interface of controller of host centre for RAID 0 and 1 on the same thing led. Including the support of Native Command Queuing and hot exchanges of drive of catch.

Ultra Ata/100

The controller for devices of PATA like the hard drives, DVD- and Cd-lead.

Ports of USB

ports of USB 2.0 Safety-speed.

acoustics of 7.1 channels

Option for the healthy device integrated with the edge, Digitals dolby and the SDR.

Modem AC97

Integrated modem.

Ethernet

Integrated 10/100 network of put-band
controller.

Diagram 192. Functions chosen in a bridge of the modern south.

But a certain number of traditional controllers for the reader with floppy disc etc are absent. Those can be placed in the superb controller of input-output.

Superb input-output

The so-called superb controller of input-output is found on all the mother charts. It is a piece of assistance, which deals with certain number less demanding functions of controller for the bridge of the south.

The tasks of the superb controller of input-output are completely well defined: that is thus perfect to employ a piece produced in series and standard, which can be employed on practically any mother chart. This helps of the prices of subsistence to the bottom, and these superb controllers of input-output come from the companies like Winbond and ITE.

Diagram 193. A superb controller of input-output of Winbond.

The superb controller of input-output is connected to the bridge of the south using a bus at fallback speed special, and this architecture is called the LPC (low account of Pine). The piece is a piece of 128 terminals containing these components:

Controller

Description

FDC

controller 82077 with compatible floppy disc

UART

Carry series, compatible controller 16550, including IrDA (infra-red)

LPT

Parallel wearing of printer

Play

For a handle

KBD

controller of keyboard 8042 compatible

PS/2

Wearing of mouse

LED

Order the small LED in the box

Ventilator

Order the ventilator

Diagram 194. Functions of which the superb piece of input-output takes care typically.

You will find many various provisions of the mother chart. The bridge of the south can be connected to other pieces of controller. �This could be:


  • ��������� A controller of RAID (for example of the promise or Highpoint, to see fig. 228, or the controller of SCSI (of Adaptec).

  • ��������� An operational processor (for example of creator or Yamaha).

  • ��������� A controller of Ethernet.

  • ��������� A controller of cable of fire, and much more.

    It belongs to the various board manufacturers mother to decide how they will employ the tiddlywinks. The bridge of the south offers loads of the equipment which can be applied so desired.

    For example, some bridges of the south have the integrated primitive healthy devices (based on standard AC-97). This uses the central processing unit of treatment to treat the noise. On some mother charts this functionality will be put at the use, whereas other manufacturers choose to integrate �a true� healthy piece, for example, of creator.

    In any case, some users will choose to install a separate healthy chart with good more functionality - independently of which noise comprises the offers of mother chart.

    Diagram 195. A situation as a whole of the many equipment of input-output offered by the bridge of the south.

    Periodic and parallel ports

    The periodic and parallel ports are included in the system of input-output, and their controller is located in the superb controller of input-output.

    The PC is equipped at least (usually two) with the doors series asynchronous. They are the male connectors on the back of the PC, which has 9 or 25 terminals, of which often only 3-4 are employed for the serial transmission.

    The parallel port is also called the wearing of printer, because is it for what it is normally employed. The connector of printer on the back of the PC is a connector female of 25 terminals.

    Diagram 196. In the medium you see the red parallel port and the two green doors series.

    The ports are also called COM1, COM2 and the LPT. They are so-called names of logical Apparatus, which the programme of starting of PC automatically assigns with these devices during starting, just like a:, C: , the SWINDLE, the PRN and the KBD are logical names.

    Diagram 197. The logical name, COM1, refer to a periodic port of input-output.

    Today the periodic doors series and devices are used much less than they were. It is particularly due to the bus of USB, which can connect the equivalent devices. But if you use, for example, a modem based by USB, it you will comprise just as if it were a normal periodic device. Below the ISDN an adaptor is installed, which is connected via the USB. It creates two virtual ports of COM which can be seen using the director of device. In this way the program commutated of Internet can use the device of USB in the same way as if it were connected to the port (physical) of COM of a:

    Diagram 198. Ports COM3 and COM9 are really a device of USB.

    Asynchronous transmission

    The doors series function using the asynchronous transmission, which it could be interesting to know a little.

    The asynchronous transmission means that data can be transferred to irregular intervals, as you could already know about use a modem. The transmission starts only when the receiver is ready to receive. And the receiver can spend unlimited time� to digest� the data. The opposite principle is synchronous transmission, where a divided clock order the step.

    The irregularity means that you must be able to establish when to send data, and when part of data was sent completely. This is carried out using a protocol, which define the lengths of package, etc, on which the shipper and the receiver must agree.

    When the shipper does not have anything to say, a jet of 1 is sent. To the beginning of a transmission, one 0 is sent. After this starting bit, the data are introduced the parts of 5,6,7 or 8 bits. Then follows 0,1 or 2 so-called bits of parity, which are used to carry out a simple control on the transfer, and the package is closed with 1 or 2 bits of arre4t. The most common standard is 8 bits of information, no bit of parity and 1 bit of arre4t. This system is often called the 8/N/1.

    Diagram 199. In a periodic transfer, the data are packed in the small, various parts, which are sent after the other.

    The� packing� necessary of the data in parts for the serial transmission takes place in the piece of UART, which belongs to the superb controller of input-output.

    The UART represents the transmitting receiver/asynchronous universal. The piece of UART converts the train of data of data between the parallel bus of input-output and periodic connection. Bytes of the bus of input-output must �be crossed upwards� in the series of little and �framed� with bits of arre4t, etc, and this takes place in the UART.

    Diagram 200. The controller of UART remballe of the data in order to be able to carry out a periodic transfer.

  •