<!–#set var="article_header" value="Plastic Surgery:
Releasing The Athlon XP To Hit 2000+” –>
Socket A Radiates
Removing the Overclocking Protection AMD engineers should give themselves a big pat on the back – the new Athlon XP/MP with the Palomino core is an excellent example of the high-quality work of which the chip manufacturer is capable. As proof of this, connecting the L1 contacts using a simple pencil, which was possible on the previous Athlons with the Thunderbird cores, is no longer possible on the new processor. This has probably foiled the plans of some hard-core overclockers out there, who start scheming about how to overclock their processor to the limit before they’ve even bought it.
Now that the Palomino is on the scene, there are additional laser locks to contend with, in addition to the new L contacts. The locks prevent you from bonding the contacts (for example, with a quick flick of a pencil or a fine brush) in order to remove the overclocking protection. But, in terms of the engineering that went into each chip, there’s no difference in the overclocking protection for the old Athlon and the current Athlon XP/MP.
And while there might be some technical details we discovered while carrying out this test, all that you really need to do is to connect the L1 contacts. This will unlock the multiplier, which is encoded at the factory. AMD encodes its chips on the L3 and L4 contacts. For more information, take a look at our tables and illustrations for this test.
Socket A Radiates, Continued
After we connected its L1 contacts, an AMD Athlon XP 1900+ has no problems running soundly at 1666 MHz (2000+).
To sum up the past few days: after a host of failed attempts and an infinite number of feedback from you our readers, we managed to prepare a clear, step-by-step guide so that all you ambitious PC users can crack the Athlon XP processor. And that’s not all – in addition to pictures and diagrams, we also show benchmark results that demonstrate the jump in the performance of the Athlon XP/MP.
The time needed for the whole shebang is about 30 minutes – at which point, the processor can be overclocked by increasing the multiplier. We dismiss the option of increasing the front-side bus, because a higher clock speed on AGP and PCI buses may plunge some components into instability.
System boot-up with the overclocked Athlon XP: The BIOS displays “Athlon XP 2000+,” a performance rating that AMD won’t even be offering for another 6 weeks or so.
Step By Step
Before you get started, make sure that you can adjust your multiplier, either in BIOS, or by switching the jumper/DIP switch settings on the board (this is the case for most Socket 462 motherboards with VIA KT133A, VIA KT266A or SiS 735 chipsets). In our test to connect the L1 contacts, we used several Athlon XP processors. For this kind of test, the Epox EP-8KHA+, which allows you to set the clock multiplier in BIOS, is an excellent basis.
You’ll need the following tools to connect the L contacts:
- Conductive silver lacquer, which is what actually connects the contacts. You’ll have to look around to find it on the Web. Try here
- Tape to insulate and separate
- Super Glue or a similar glue to fill the laser locks
- Scalpel to remove any remaining glue
- Multimeter to determine the resistance
View is of an Athlon XP 1900+. The arrow shows the L1 contacts that will be modified later.
Why The Pencil Trick Doesn’t Work
Unlike the conventional Athlon (ceramic substrate with a Thunderbird core), on which the L1 contacts can be connected with a pencil stroke, AMD has put more work into protecting the Palomino’s multiplier. While the resistance between earth and L1 (bottom row) approached ‘infinity’ on the old Thunderbird Athlon, we measured a resistance of 945 Ohm (about 1 k() on the Athlon XP (Palomino core, organic substrate).
These measurements show why the pencil trick won’t work: if the L1 bridge was closed with a pencil stroke, the resistance provided by the graphite would still be too high. The internal pull down resistance ratchets the signal level down so far that the CPU would never interpret the contact as a closed bridge. In other words, AMD knew that its new feature would hinder overclockers. The only way around it is to use a substance with a minimal contact resistance, namely, conductive silver lacquer, which you can get at an electronics store.
The pull down resistance between a grounded point and the L1 contacts stands at about 1 Kilo Ohm – the pencil trick is a no-go!
An old Athlon with a Thunderbird core: the resistance in the L1 contacts, which were connected using a pencil, is higher than 1 Kilo Ohm – however, the CPU still doesn’t have an internal pull down resistance.
Another measurement showed that the direct contacts on L1, L3 and the triangle on the left (in blue) are grounded. Under no circumstances should these points be connected to the L1 bridges with the silver lacquer – it would cause every attempt at closing the bridges to fail.
Here’s The Trick – Masking The Contacts
Before fiddling around with the conductive lacquer, the “burn holes” created by AMD’s laser separation have to be sealed. If the silver lacquer trickles into these holes, you’ll be up against the same grounding problem mentioned earlier. To the naked eye, it’s not obvious that there’s grounded copper leaf underneath.
First, the L1 contacts (top and bottom rows) are taped off using Scotch tape or another transparent tape. The idea behind this is to allow you to insulate the laser locks using super glue in the next step.
View of the L1 contacts on an Athlon XP 1900+.
Extreme close-up of the L1 contacts.
Careful! Press down firmly on the tape so that the glue won’t go where it’s not supposed to go.
Applying The Super Glue – Insulating The Laser Locks
Once the contacts have been fully insulated with tape, the super glue can be applied to the gap. Make sure to apply a well-measured dose of the glue so that only a modicum of glue ends up on the processor.
Applying super glue to the taped-off gap between the L1 contacts.
Close-up of the laser locks once filled with glue.
Removing The Tape And Remaining Glue
The super glue needs about 10 minutes to dry completely after it has been applied. Then, carefully remove the tape and use the scalpel to cautiously scrape away any excess glue.
Removing the excess glue between the L1 contacts with a scalpel.
A Second Masking: Applying Conductive Silver Lacquer To The L1 Contacts
Now it’s time to neatly connect the separate L1 contacts (the top one with the bottom one) using the conductive silver lacquer. You have to mask the contacts with scotch tape again for this procedure; otherwise, the connection won’t be made properly. First, apply the tape lengthwise along each L1 contact (In the picture shown from top to bottom). Then, make a masking window by applying the other two strips of scotch tape crosswise (shown in the picture, from left to right). With a slew of failed attempts and destroyed processors under our belt, we would like to urge every user to follow our instructions!
Every individual contact group is taped off separately, so that the silver lacquer can be applied specifically. The picture above shows how precisely every contact needs to be taped off. Otherwise, it’s impossible to connect the contacts adequately. Then, apply the conductive lacquer with a small brush.
Conductive silver lacquer, which can be found at Conrad Electronic, or in any electronics store.
Applying the lacquer to the masked-off window. The window should literally be filled with the conductive lacquer.
Close-up of the first contact connected with silver lacquer.
Now, all you have to do it remove the tape, and there you will have an L1 contact that has been connected relatively cleanly. Follow this procedure with all remaining contacts until all L1 bridges have been closed with the conductive lacquer. Now, test all the connections once through (bottom side with top side). The resistance should be at about 0 Ohms! You should also double-check that there aren’t any cross-connections between the individual rows. If you discover a cross-connection, sever it carefully, using the scalpel or a razor blade. When measuring, make sure that you don’t press down too hard with the test prods, or you might rub away the conductive lacquer. This procedure is also reversible. All you need to do is to remove the bridges using a stiff eraser, and, if desired, start again from the beginning.
Test: Athlon XP 1900+ Overclocked At 2000+
All the contacts have been properly connected – to keep them that way you can also apply a strip of tape on top of the connections. We installed the processor in our test board, an Epox EP-8KHA+ with a VIA KT266A chipset. The following images are proof that the multiplier is now freely adjustable.
The multiplier is now freely adjustable in the BIOS menu.
The multiplier 12.5X isn’t available on our board – the CPU interprets 13X as the field value. This reading is incorrect and should be taken care of by the BIOS specialists at Epox.
Modifying the core voltage in BIOS to overclock the processor.
To pump the Athlon XP 1900+ up to 2000+, we had to raise the core voltage to 1.85 Volts.
A display with the new clock speed and multiplier setting under Windows 98. After BIOS shows the Athlon XP as having a 1666 MHz clock speed (Athlon XP 2000+), you can boot up your operating system (Windows 98 SE in this case). The popular tool, WCPUID, offers the following data: core clock speed 1666 MHz, multiplier 12.5X and front side-bus speed of 133 MHz. Made it!
Checking the new clock speed and multiplier setting under Windows XP.
Multiplier And Voltage Settings
We set up the following two tables for users who want to have the encoding settings for the multiplier and the CPU core voltages conveniently at hand.
Encoding settings for the clock multiplier.
If you own a motherboard geared toward overclockers (i.e., one that allows you to alter the multiplier in BIOS), closing all the L1 bridges is still the most comfortable solution. That was why we’ve only demonstrated this method here – if all your L1 bridges are closed (unlocked), setting a free multiplier in BIOS is a piece of cake. The CPU originally comes with all its L1 bridges open (locked). AMD sets the multiplier on L3 and L4. However, if you start fiddling around with L3 and L4, you run the risk that you won’t be able to undo your new multiplier. For this reason, THG has decided to refrain from offering any solutions for L3 and L4.
Encoding settings for the CPU core voltage on all AMD Athlon XP/MP processors using the L1 bridges.
Motherboards geared toward overclockers allow you to adjust the core voltage manually. Anyone in possession of a motherboard that only offers automatic voltage recognition will generally have to increase the original core voltage in order to overclock the CPU properly.
False Steps
Before finding out which method was the best, we went through a series of failed attempts in our tests. One of the main issues has to do with masking the individual contacts. Our initial attempts were made using paper, which is completely unsuitable for use with conductive silver lacquer. There’s no guarantee that the CPU substrate surface has been completely covered. If you drip the lacquer into a homemade masking window made of paper, it will get underneath the paper and smear the CPU surface, and all your work will go right down the drain.
Failed attempt to mask the L1 contacts by using strips of paper.
Close-up of L1 bridges that haven’t been cleanly closed. Other mistakes in the following pictures:
Doesn’t work – the pencil trick won’t work with the Athlon XP/MP.
The graphite marks made by a pencil, magnified. The resistance provided by the graphite is too high – the connections are ineffective! For the techies out there – this kind of graphite mark generally carries a contact resistance exceeding 1 k(. This level of resistance can’t generate the necessary closed signal level due to the internal pull down resistance of the Athlon XP. The old Thunderbird Athlon doesn’t have this pull down, which is why the classic pencil trick still works on it. If you don’t mask neatly enough, you might cause the same problems that are portrayed in the images below.
In this image, the glue layer intended just for the laser locks has seeped out too far into the area where the contacts are.
Close-up of a scraped glue layer.
Testing Procedures And Notes
| Intel Hardware Socket 478 |
|
| Processor | Intel Pentium 4/2000 MHz (400 MHz QDR FSB) Intel Pentium 4/1900 MHz (400 MHz QDR FSB) Intel Pentium 4/1800 MHz (400 MHz QDR FSB) Intel Pentium 4/1700 MHz (400 MHz QDR FSB) Intel Pentium 4/1600 MHz (400 MHz QDR FSB) Intel Pentium 4/1500 MHz (400 MHz QDR FSB) Intel Pentium 4/1400 MHz (400 MHz QDR FSB) |
| Motherboard | ASUS P4T-E (I850) Revision: 1.00 |
| Memory | 2 x 128 MB, RDRAM, 400 MHz, Viking |
| AMD Hardware Socket 462 |
|
| Processor | AMD Athlon XP 1900+ OC @ 2000+ (1666/266 MHz DDR FSB) AMD Athlon XP 1900+ MHz (1600/266 MHZ DDR) AMD Athlon XP 1800+ MHz (1533/266 MHZ DDR) AMD Athlon XP 1700+ MHz (1467/266 MHZ DDR) AMD Athlon XP 1600+ MHz (1400/266 MHZ DDR) AMD Athlon XP 1500+ MHz (1333/266 MHZ DDR) AMD Athlon 1400 MHz (1400/266 MHZ DDR) |
| Motherboard | Epox EP-8KHA+ (VIA KT266A) Revision: 2.0 |
| Memory | 256 MB DDR-SDRAM, CL2, PC2100, Micron |
| General Hardware | |
| Graphics card | GeForce 3 Memory: 64 MB DDR-SDRAM Memory clock: 400 MHz Chip speed: 250 MHz |
| Hard drive | 40 GB, 5T040H4, Maxtor UDMA100 7200 rpm 2 MB Cache |
| Drivers & Software | |
| Graphics card driver | Detonator 4 Serie V21.85 |
| DirectX version | 8.1 |
| Operating system | Windows XP Final, Build 2600 (Englisch) |
| Benchmarks & Settings | |
| Quake III Arena | Retail Version 1.16 command line = +set cd_nocd 1 +set s_initsound 0 Graphics detail set to ‘Normal’ Benchmark using ‘Q3DEMO1’ |
| 3DMark2000 | Version 1.1 Build 340 – default Benchmark |
| 3DMark2001 | Build 200 – default Benchmark |
| SiSoft Sandra 2001 | Professional Version 2001.3.7.50 |
| Newtek Lightwave | Rendering Bench SKULL_HEAD_NEWEST.LWS |
| mpeg4 encoding | Flask V0.6 (MPEG 3) DivX codec 4.02b codec Compression: 100 Data Rate: 1500 Kbit 720×480 Pixel, 25 fps no Audio |
| Studio 7 | Version 7.02.7 (MPEG 2) |
| Sysmark 2001 | Patch 3 |
| Lame | Lame 3.89 MMX, SSE, SSE 2, 3DNow |
| WinACE | 2.04, 178 MB Wave file, best compression, Dictonary 4096 KB |
| CINEMA 4D XL R6 | CineBench 6.103 |
| Suse Linux 7.3 | Kernel 2.4.13 Compiling |
Benchmark Results For The Athlon XP 2000+
We performed a total of 19 different benchmark tests in order to obtain the most complete, well-balanced view of how the Athlon XP 2000+ performs. We determined OpenGL performance using four different Quake tests – Direct3D performance from the DirectX package is determined using the 3D Mark 2000 (based on DirectX 7) and the 3D Mark 2001 (based on DirectX 8). The different MPEG-encoding benchmarks portray a comprehensive testing scenario – the Lame MP3 Encoder was used to encode a 178 MB WAV file into MPEG-1 Layer 3 format. Still a classic, our MPEG-4 test converts a file from a commercial DVD-ROM into MPEG-4 format using Flask Mpeg and Divx. A new addition to our benchmark suite is encoding an MPEG-2 file within a project by using the video editing software Pinnacle Studio 7.
OpenGL Performance: Quake 3 Arena
In both time-demo runs of Quake 3 Arena, the Intel Pentium 4 processor clocked at 2000 MHz came in just a tad ahead of the AMD Athlon XP 2000+ (1666 MHz). The Athlon XP 2000+ beat out the competition in the NV15 demos.
Direct3D Performance – DirectX 7: 3D Mark 2000
The 3D Mark 2000 determines DirectX 7’s Direct3D performance under Windows XP. The diagram shows that the Athlon XP 2000+ has quite a jump on the other processors – it even scores 1300 points higher than the Intel Pentium 4/2000!
Direct3D Performance – DirectX 8: 3D Mark 2001
The 3D Mark 2001 determines DirectX 8’s Direct3D performance under Windows XP. The Athlon XP 2000+ has an edge over the fastest Intel Pentium 4, but it isn’t as stunning as it was in the 3D Mark 2000 and DirectX 7.
MP3 Audio Encoding: Lame MP3
The Lame MP3 Encoder under Windows XP is used to convert a 178 MB sound file from a WAV format to an MPEG-1 Layer 3 format. The AMD Athlon XP 2000+ is the top dog in this race, encoding the file 15 seconds faster than the Intel Pentium 4/2000 does.
MPEG-4 Video Encoding: Flash MPEG and Divx
The Athlon XP 2000+ hits a refresh rate of 24.5 frames per second while encoding – this widens its lead over the Intel Pentium 4 even more!
SiSoft Sandra Benchmarks: CPU and Multimedia
In the SiSoft Sandra Benchmark 2001, we see that, with the exception of the memory benchmark, of which the Intel Pentium 4/2000 is still the king, the AMD Athlon XP 2000+ is the overall winner. A small glitch continues to crop up, causing the AMD Athlon 1400 to look better than the new XP processors.
3D Rendering: Newtek Lightwave
The Lightwave benchmark shows an entirely different picture: The Athlon XP 2000+ is somewhat faster than an Intel Pentium 4/1400.
Office Performance: Sysmark 2001
The Athlon XP 2000+ takes first place in the category “Unofficial Score.” The Athlon XP 2000+ is a hair’s breadth away from topping 200 points. Athlon’s lead is even wider in the Office Performance test – all Athlon processors come out ahead of the Intel CPUs.
Compiling Linux: Suse Linux 7.3 / Kernel 2.4.13
When it comes to compiling the latest Linux kernel, the AMD Athlon XP 2000+ is the fastest CPU. It needs a mere 202 seconds, whereas the Intel Pentium 4/2000 needs 264 seconds to compile the same kernel. That’s quite a difference!
Archiving: WinACE 2.04
Archiving is a very practical application. WinACE 2.04 was used under Windows XP to archive a 178 MB WAV file while the clock was running. Now, the Athlon XP 2000+ takes the lead in this discipline.
3D Rendering Performance: SPECviewperf ‘Lightscape’
In the Lightscape benchmark, the Intel Pentium 4/2000 continues to be a nose ahead of the competition, while the fastest Athlon XP 2000+ takes third place.
MPEG-2 Video Encoding: Pinnacle Studio 7
The Athlon XP 2000+ was the fastest at creating an MPEG-2 film using Pinnacle Studio 7. That’s something, considering that the software is optimized for use with the Intel Pentium 4!
Conclusion: Athlon XP 2000+ Performance Already Available
Our guide will definitely help you to unlock the clock multiplier on any AMD Athlon XP/MP. Provided that you have the right tools and accessories, you can use conductive silver lacquer to connect L1 contacts on an Athlon with a Palomino core.
After modifying the contacts, you’re free to set the multiplier in BIOS or on a DIP switch. However, the maximum setting is currently limited to 12.5X, which allows you to reach a clock speed of 1666 MHz (12.5 X 133 MHz = 1666 MHz) without having to increase the front side-bus clock speed. This guide has shown you how to obtain a powerful Athlon XP 2000+, a processor that AMD won’t even officially introduce for another 6 weeks.
If you’re still hungry for more, you can eke the very last spark of performance out of your system by gradually increasing the FSB speed. By setting the multiplier to 12.0X and pumping the FSB speed to 153 MHz, we managed to overclock the Athlon XP 1900+ (clocked at 1600 MHz) to 1836 MHz (12.0 X 153 MHz = 1836 MHz). This clock speed might be equivalent to a performance rating of 2200+. However, we only managed to reach such a high clock speed by using a perfectly adjusted water-cooling system. Anyone planning to reach this kind of speed with a standard cooler is daydreaming!
Here’s an example to illustrate this fact: even when we used a powerful Silverado cooler or the Swiftech monster (MC462), it was still impossible to stabilize the system at 1836 MHz.
To reach lightning-fast clock speeds such as 1836 MHz, a powerful water-cooling system is an absolute must.
CPU temperature of 29 degrees Celsius with our water cooler.
In addition to our instructions and hands-on tips, we’ve offered an exclusive benchmark test of the Athlon XP 2000+, which isn’t slated to be introduced by AMD until January 2002. The diagrams clearly prove that the Athlon XP 2000+ beats its arch-rival, the Intel Pentium 4/2000 in all the disciplines. It now remains to be seen which of the two manufacturers will take the first step and introduce the next CPU generation.
Please follow-up by reading Второе THG видео: Разгон Athlon XP/MP
