Yes, you heard it right, I may not be PimpRig staff but whose to say I can't write an awesome review. I always was one of the top writers in high school so let's see if I still got it......

How I came across this processor....

As Christmas time rolls around, hardware gets cheaper as computer store chains lower their prices and have sales on certain items. As a result, people give the gift of money for Christmas, and I ended up with 83 dollars to blow on something for my PC. At least my Half-Life server. While I was testing some Athlon boards I nabbed at my other job for free, one ASUS A7S-VM ate my server processor, which was a Duron 1300 'Morgan Core'. It was so hot you could have fried eggs on the die if it was big enough.

When I got my Christmas money, the first thing I did was run to the local computer store, Bits Computers, to get a replacement Duron 1300. But they were out of stock. Instead, for the price of my old 1300 Morgan, I got a 1400 'Applebred Core', also referred to as the 'Thornton Core'. Not a bad deal for $42.40.

The Processor

Well at first I thought he gave me an Athlon. But it said 'Duron 1400' on the chip. It seems that AMD has switched from using a ceramic die on the Duron to PC board. And for all you overclockers out there, the L1 bridge is still visible on the die and is just begging to be shorted out with a pencil or a conductive pen.

Below are some shots I took of the processor. I bought it OEM.





There is a slight noticeable difference in cooling, it seems to run a bit cooler than the ceramic 1300 die, with a stable 30-min BIOS idle temp. of 37*C, as opposed to my 1300's 42*C idle temp. With an ambient temp. of about 71*F, I installed the chip into the server. The specs for the server are listed below.

Motherboard: DFI AZ-30 TL mATX
CPU: Duron 1.4 GHz 'Applebred/Thornton Core'
CPU Fan Setup: Generic COMPUSA Copper Heatsink, 60MM 7100 rpm Delta 38CFM screamer, Jet Art thermal paste.
Memory: 1X 512MB DDR400 R-Data @ DDR266 speeds, PNY copper heat spreader
Video: Onboard S3 Savage

And this is what it looks like with the side panel off.



As you can see, my server is nothing special. But it works and it works well.

Installation and BIOS Setup

I'm not sure if it was a multiplier limit on the motherboard or what, but it seems that the new Applebred won't run at 100 MHz FSB unlike its ceramic predecessor, the Morgan Core. On the DFI it posted as 'Unknown CPU Type/1050 MHz'. So I'm thinking, 'okay, I know this is a 1400, and 100 X 10.5 would be 1050 MHz. But it ran at 13 X 100 so it wasn't a matter of multipliers'. The DFI has a jumper on the board that forces a 100MHz or a 133 MHz FSB. So I set it to 133 MHz FSB, powered it up and this time it posted as 'Unknown CPU Type/+1600' Now I'm thinking 'Well I really did get an Athlon.' The 1600 if I remember correctly runs at 1.47 GHz. But I went into the BIOS and the CPU speed read '1400 MHz', 133MHz FSB. So the CPU speed problems are fixed.

I aslo tried to see what it would do in an older motherboard with a KT133 chipset like my Gigabyte GA-7IXEH. I must say I was a bit disappointed. It wouldnt post. So I cleared the CMOS and tried again and it posted at 1050 MHz, but it didnt post after I hit the restart switch. Here are the specs on this rig below.

Motherboard: Gigabyte GA-7IXEH Rev. 1.0 w/5.0 BIOS Flash
CPU: AMD Duron 1.4 GHz 'Applebred Core'
CPU Fan Setup: Thermaltake Volcano 6 Cu w/80MM Vantec Tornado Fan, AS Ceramique
RAM: 1X 512MB PC133 Kingston, 2X 256MB PC133 Infinieon, total 1 GB.
Video: 3DFX Voodoo 5 5500 AGP 4X, 64MB, dual VSA-100 166MHz processors.

It seems the Applebred wont run on 100 MHz FSB. This is my biggest disappointment aside from the little CPU speed setups because the majority of people who buy Duron processors install them on older motherboards. Think about it, why buy an expensive board that costs 130 plus and get a 40 dollar processor? Chaintech Apogee on a Duron 950.

Unless you have a board that can force 133MHz FSB by jumpers on the board itself I would advise you to steer clear of the Applebred if you have a KT133 or an older Socket A motherboard. Stick with the Morgan if you can't force 100/133MHz FSB. So it's back to the DFI Server rig for testing purposes.

I also noticed the difference in core size between the Morgan and Applebred. The Applebred uses a smaller core compared to the Morgan, which most likely accounts for it's lower idle temps. Now that I have allowed temps to stabilize in BIOS, its time to boot into XP Pro and see how it runs applications (like a Half-life server)

Running the Applebred in a Windows Environment

Boot-up time was about the same as the 1300. No real large difference in speed, however the biggest difference between the 1300 and 1400 is the cooling. Unfortunately I was hoping to do the review on the Gigabyte rig because Motherboard Monitor 5 wont work with my DFI board, the temp. settings are way off and everything comes up way inaccurate, the CPU temp is listed as -93*C and all the voltages are astronomically high and low.

The server also runs Folding on it, and after 5 min. it had already chewed through 30/250 of the WU it was folding at the time. During a map called Killbox on the HL server, a whole bunch of Apple Clan members joined the server, and were doing things that would normally lag or crash the server, like throwing tons of snarks and planting lots of tripmines, usually after about 13 players the server starts to die, it lags up and just keels over. But 16 people were on it, the server was at its maximum number of rated players and it was doing just fine, no lag or loss in FPS. Maintained 100 FPS the whole time, not dropping below 95 even at its limits. Very impressive for something that cost 42 dollars.

To wrap it all up, if you have a newer entry-level motherboard like a DFI or Mach Speed, the 1.4 'Applebred' is your hot ticket if you are budget-conscious. But if you have an older Socket A KT133 chipset board that can't force 133MHz FSB by jumper on the motherboard then stick with the 'Morgan'

Overall I give it 7 out of 10

Feel free to post your comments in this thread.

bump for those who havent seen it yet

nice review, lots of info

good job!!!

more bumpage

Quote: Originally Posted by j-dogg more bumpage

You like your bumpage don't you 'dogg?

It's a great article as far as it goes: what it lacks is empirical comparisons - i.e. before & after benchmarks.

yah i know but my 1300 was dead before i could benchmark it

Nice article there, but as mentioned above it is lacking a few things. What about overclocking the CPU?

dogg,
I'll give ya an A for effort, you managed to get in all the "zip-words" that one would read in an article.
You need to simplify your findings, I.E. Temps gathered by a standard means, and use either C or F or both, but not one of each, to record them.
Any temp that is reported as "about", like about 71f, isnt but a guess, and really shoots your accuracy curve down.
The use of graphs helps to tell of your findings, less confusing too.
I was reading about a CPU, then I was reading about motherboards.
It was a CPU review right? If the platform that you are using doesn't have the correct functions for a review...you have no review.

I wasn't gonna say anything about your servers' guts, I think you mentioned, "it works, and works well"...
But I just can't resist,
WTF is all that stuff, wadded up behind the drives? Dood there is no excuse for all that. Take some time, and at least a few zip-ties and untangle that mess...
I've had plates of spagetti and meatballs, that were more organized...not to mention the air-flow blockage...

Not trying to talk smack, just a couple of hints...Good job otherwise.

Now you can really appreciate all the work that goes into what seems like a simple review. Gotta give the STAFF kudos for the time and effort, they put out for us.

Now do the L12 mod, and open up the cache.

Never thought about overclocking, I don't have anything hooked up or volt-modded yet. I just got it so I could get my server back up and running. But I might get another one in the future and play with it some

Its just the multis on both of my boards are locked. Had I got it to do something in my Gigabyte then we could have played with it some

I will agree not too bad for my first. It is missing some stuff though. Great comments keep it coming.

Oh yah and Zenn, the server isn't a form thing its a function thing. I really dont care what it looks like inside as long as it doesn't get over 50*C and everything works. Now my Giga-byte, that's a whole 'nother story and man it needs some cleaning.

I have one of these 1.4's in friends rig I built on the cheap, at 2ghz plus on a ASUS A7V333 Mobo
Go for it!

Quote: Originally Posted by j-dogg Feel free to post your comments in this thread.

OK, here goes:

I'm going to try and give you a completely honest critique - I'm not flaming you, but as the purchaser of all computer related hardware for a large international insurance company, It's my job to read reviews pretty much all day, every day, so I have a bit of experience when it comes to what should and should not be in a product review.

First of all, big kudos to you for taking this task on. It's not an easy job to write a product review. I give you a C for effort though, because you didn't research anything. However, I thought it was a terrible review - read on to find out why.

When reading your introdution, you mention that you got a 1400 Applebred core. That's it. No other specs from AMD. I'll be honest - I've never heard of an 'Applebred core' before, so I was looking for more information on it. What I got from your article was nothing. This is what I would have expected from an introduction of a new CPU:

• Cache size
• Actual frequency
• PR Rating (if any)
• Voltage rating
• Included instruction sets (MMX, MMX2, 3DNow etc.)
• Recommended Bus speed
• New features, and what's been changed compared to the old incarnation of the product

I think you get the idea. All this information is readily available from the AMD web site, and found with little effort.

Moving on...

One thing about your article that really annoyed me (and made me question whether you should have written this at all) is all the generalizations you make. A review needs to have accurate and specific information. Case in point: (I've bolded my concerns)

Quote: Originally Posted by j-dogg There is a slight noticeable difference in cooling, it seems to run a bit cooler than the ceramic 1300 die, with a stable 30-min BIOS idle temp. of 37*C, as opposed to my 1300's 42*C idle temp. With an ambient temp. of about 71*F, I installed the chip into the server.

1. You previously mentioned that you have 2 platforms (motherboards) to test this CPU on, however you neglected to mention which of the motherboards you're getting these temps from.
2. A 'slight noticable difference'? Which one is it? Slight or noticable?
3. ABOUT 71*F??? There is no place for "about" in a review. It MUST BE specific.
4. You just finished listing all of these temperatures, then say "ith an ambient temp. of about 71*F, I installed the chip into the server." WTF??? Where did you get all of these teperatures from if it wasn't installed into the machine?

The next area I have a concern with is here:

Quote: Originally Posted by j-dogg I'm not sure if it was a multiplier limit on the motherboard or what, but it seems that the new Applebred won't run at 100 MHz FSB unlike its ceramic predecessor, the Morgan Core. On the DFI it posted as 'Unknown CPU Type/1050 MHz'. So I'm thinking, 'okay, I know this is a 1400, and 100 X 10.5 would be 1050 MHz. But it ran at 13 X 100 so it wasn't a matter of multipliers'. The DFI has a jumper on the board that forces a 100MHz or a 133 MHz FSB. So I set it to 133 MHz FSB, powered it up and this time it posted as 'Unknown CPU Type/+1600' Now I'm thinking 'Well I really did get an Athlon.' The 1600 if I remember correctly runs at 1.47 GHz. But I went into the BIOS and the CPU speed read '1400 MHz', 133MHz FSB. So the CPU speed problems are fixed.

1. 'It seems that the new Applebred won't run at 100MHz'. No, it won't. If you had looked at the specs of the processor in the beginning, you would notice that it's designed to run on a 266MHz DDR FSB. (133MHz)
2. 'On the DFI it posted as 'Unknown CPU Type/1050 MHz'.' To me, this indicates one of 2 things: a) you need a BIOS flash; or b) this motherboard does not support this processor.
3. 'So I set it to 133 MHz FSB, powered it up and this time it posted as 'Unknown CPU Type/+1600' This definitely confirms my point #2 directly above.
4. 'But I went into the BIOS and the CPU speed read '1400 MHz', 133MHz FSB. So the CPU speed problems are fixed.' Are you sure? You're 'testing' this processor on a downlevel BIOS, possibly on an unsupported board. Just because the frequency appears to be correct does not mean that the 'CPU problems are fixed'.

Quote: Originally Posted by j-dogg I also noticed the difference in core size between the Morgan and Applebred. The Applebred uses a smaller core compared to the Morgan, which most likely accounts for it's lower idle temps.

Again, how do you know?? What about the fact that this processor runs at 1.5v compared to 1.75v on the older version? Might that be the cause of the lower temperatures?

Quote: Originally Posted by j-dogg Boot-up time was about the same as the 1300. No real large difference in speed, however the biggest difference between the 1300 and 1400 is the cooling. Unfortunately I was hoping to do the review on the Gigabyte rig because Motherboard Monitor 5 wont work with my DFI board, the temp. settings are way off and everything comes up way inaccurate, the CPU temp is listed as -93*C and all the voltages are astronomically high and low.

I have a HUGE issue with this paragraph. It begins with another generization: "Boot-up time was about the same as the 1300. No real large difference in speed". Then, in the same sentence, you move to another topic, which you've already covered: "however the biggest difference between the 1300 and 1400 is the cooling." Now this is the part that REALLY confuses the hell outta me. You mention earlier in your article that the Gigabyte board you have will not run with this new CPU. Fine. Then you go on to say that any information you get from your DFI board with MM5 is highly innacurate??? WTF??? This throws any credibility you had left right out the window.

In the next paragraph, you get to your 'testing'. In 5 minutes it gets through 30/250 wu?? My P42.6c with 1GB of RAM can barely get through 2/250 wu's in 5 minutes! This is highly suspect. My theory: The previous CPU had crunched through 29 of those 30, and the new CPU was picking up where the old one left off.

The next 'test' is your Half Life server, and the fact that it had 16 people on it when the old one usually dies at 13 users. That's all fine and dandy, but there are a lot of things out of your control that basically invalidate this test - bandwidth, player participation, map and activity to mention a few. You also have no idea if the other CPU you were using was damaged in any way, due to age, full-time 100% CPU usage, overclocking and god knows what else you've done to it.

You then proceed to give the new CPU a 7/10. Based on what? The fact that your server didn't crash and the idle temperatures are better?

For what's it's worth, you did not write a review. You wrote a commentary on your experience of installing a CPU into your existing rig - nothing more.

I hope you can take this the right way...you're a nice guy, and somewhat helpful on the forums. I was going to PM this to you, but I figured since you posted in the forum and asked for feedback, I would give you that feedback via the forum.

All in all, these are my suggestions for your next foray into product reviews:

• Do lots of research, and credit your sources.
• Read lots of reviews from top-rated hardware sites (including PR) and take note of how they structure their reviews
• Accuracy is paramount! Generalizations have NO PLACE in a review.
• When benchmarking, ensure consistency - preferably a fresh OS install, and run all benchmarks at default settings. Do exactly the same thing for each product you review, and try to perform your tests so that others can duplicate your results.
• More pics, and if you can, make them a bit clearer - it's difficult to see the chip, and the PC pic is really out of focus.

Good luck on your next review.

I woulldnt mind overclocking my CPU, and overclocking my 3D card also, just a bit, but not too much because i am not water cooled yet. I have a ATX TX-1 case on order also so i wont do any overclocking until i get the internals transfered to the new one. Do you go into the bios to overclock?

Quote: Originally Posted by JimsZR2 I woulldnt mind overclocking my CPU, and overclocking my 3D card also, just a bit, but not too much because i am not water cooled yet. I have a ATX TX-1 case on order also so i wont do any overclocking until i get the internals transfered to the new one. Do you go into the bios to overclock?

Yes, you usually need to set the CPU frequency in the Bios, unless you have a newer N-Force chipset on your motherboard, then you can Overclock it through Windows XP with Clock Gen

This was a thread on the new Duron, do you have one?

PimpJoose, good critique without being harsh.
I built a system a few months ago as stated above with the new Applebred core 1.4, on a Asus A7V33-X Motherboard, it easily overclocked to 2.2GHZ

It's basicaly a T-Bred core with 128K cache instead of 256, and no PR rating.
In the NewEgg customer reviews, they say the 1.6 version has reached 2.5ghz