"Do we have any idea what the Athlon will scale up to? We know it will scale up to 1.7GHz; it will probably scale even farther once it drops to .15u and finally to .13u. Has anyone said what the limit is for the Athlon architecture?"
The current architecture won't be scaling very high. I'd be impressed if it made it to 2GHz at all. AMD won't be using the 0.13 process until 2002.
"And beyond that, there's the Hammer family, which will probably scale even farther. Even though it's a 64-bit chip, there will probably still be a model geared towards the desktop/workstation."
The Hammer is AMD's answer to the Itanium. That's an entirely different discussion.
-Raystonn
-- The center of your digital world --
The current architecture won't be scaling very high. I'd be impressed if it made it to 2GHz at all. AMD won't be using the 0.13 process until 2002.
"And beyond that, there's the Hammer family, which will probably scale even farther. Even though it's a 64-bit chip, there will probably still be a model geared towards the desktop/workstation."
The Hammer is AMD's answer to the Itanium. That's an entirely different discussion.
-Raystonn
-- The center of your digital world --
Any piece of junk will allow me to write a resume, surf the net and send emails to my grandma. Lets forget low spec systems and get to the real issues. Hardware isn't worth a dime without software....so what does intense software require? For all of you pretenders have a look at an article in Toms Graphics Guide...written May 15 and titled Proffesional Affair. It's about the various pro graphics cards and rendering speeds etc. Check it out. A P3 beats a same speed Athlon in 99% of the tests. One more thing. Dual Athlon boards are currently unavailable and this has been the case for years. So if your a serious power user requiring multi-processing power what do you do with an Athlon? Use it as a paper weight? Make a key ring out of it? Squash ants as they walk across the pavement? Stick it up someone else's arse? Maybe stick it up your own arse if your an AMD low spec my wise and beautiful friend?
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
Kelledin
Distinguished
- Mar 1, 2001
- 2,183
- 0
- 19,780
I've never heard anyone mention any hard data on the clockspeed scaling capabilities of the Athlon. I'll need links please...we all seem to be in the habit of demanding links, so I guess I'll just be fashionable. 
And the Hammer could still very well serve as an answer to the P4, even if that wasn't its original intent. The Clawhammer variant seems to be aimed more at workstations...
Kelledin
<font color=red>"Step away from the gimp suit and put your hands on top of your head."</font color=red>
And the Hammer could still very well serve as an answer to the P4, even if that wasn't its original intent. The Clawhammer variant seems to be aimed more at workstations...
Kelledin
<font color=red>"Step away from the gimp suit and put your hands on top of your head."</font color=red>
Tonestar, as much as I agree some of what you said, your digression at the end invalidates the whole argument in the eyes of many. I don't want to censor anybody, just let you know that it actually hurts your argument rather than helping it.
-Raystonn
-- The center of your digital world --
-Raystonn
-- The center of your digital world --
My point is simply this Ray...paper specs vs real world tests can often be miles apart. Speed can also come at the compromise of stability. Simply why spend money for minor gains? And if you need major gains why buy into AMD? Its all relative unless your actually in the business of selling CPU's. Then you can go chasing links and sending out press releases. Till then keep it real.
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
>Speed can also come at the compromise of stability.
Please post evidence of lack of stability with Athlon CPUs
>Simply why spend money for minor gains?
Sounds like an argument against buying a P4 at this time.
>And if you need major gains why buy into AMD?
Nope, you get an Alpha. Or maybe even a dual alpha. But your best price/performance ratio for a single CPU system at this moment is AMD.
And by the way, SMP isn't the answer to everything. I have a bus-limited simulation that runs substantially faster on a single CPU box. If SMP is the answer for you, then yes, AMD is not a solution at this time. But the PIII isn't to enticing either.
In theory, there is no difference between theory and practice.
In practice, there is.
Please post evidence of lack of stability with Athlon CPUs
>Simply why spend money for minor gains?
Sounds like an argument against buying a P4 at this time.
>And if you need major gains why buy into AMD?
Nope, you get an Alpha. Or maybe even a dual alpha. But your best price/performance ratio for a single CPU system at this moment is AMD.
And by the way, SMP isn't the answer to everything. I have a bus-limited simulation that runs substantially faster on a single CPU box. If SMP is the answer for you, then yes, AMD is not a solution at this time. But the PIII isn't to enticing either.
In theory, there is no difference between theory and practice.
In practice, there is.
Grizely1
Splendid
- Dec 31, 2007
- 7,764
- 0
- 30,780
You are wrong. The Hammer series is not AMD's answer to Itanium. Well, the SledgeHammer is. The ClawHammer, however, is aimed at the desktop market. The only difference between the two is the cache. One for servers one for desktops.
-----------------
All your RAMBUS are belong t............ ahh screw it
-----------------
All your RAMBUS are belong t............ ahh screw it
Dual Alpha? No my dual P3 is doing just fine thank you. Did you read the article titled "Professional Affair" in Toms Graphic Guides you twit? Are you saying that in your pathetic world a P3 couldn't do what your Athlon does? If your application is bus limited then why didn't you just get an old 486 out of somebody's dumpster and just use that? Leave SMP concepts for those that understand it.
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
>Dual Alpha? No my dual P3 is doing just fine thank you.
You think a dual PIII is even vaguely comparable to a dual Alpha? And yet you call me a twit? Go find a clue, It's not even close. You can start with those SpecFP benchmarks.
Here I'll help you out:
http://www.spec.org/osg/cpu2000/results/cfp2000.html
##################################### base peak
Compaq Computer Corp AlphaServer ES40 Model 6/833 1 590 658
####### Intel Corp. Intel OR840 (1 GHz Pentium III) 1 327 335
Here are the FP rates.
This shows how well the SMP architecture scales.
########################## #CPU base peak
Compaq AlphaServer ES40 Model 6/833 1 6.84 7.63
Compaq AlphaServer ES40 Model 6/833 4 24.3 26.1
Scaling Factor 3.55 3.42
I couldn't find any Intel numbers on this page. Wonder why?
Because Intel SMP scaling sucks.
Oh, but wait ... the Alpha is only 833 MHz, it must be slower.
Yeah, right...
Sorry, it doesn't do windows though.
>Did you read the article titled "Professional Affair" in
>Toms Graphic Guides you twit?
Frankly, in my work I don't give a damn about graphics performance. Pretty pictures are for generals, the analysts look at the numbers.
>Are you saying that in your pathetic world a P3 couldn't
>do what your Athlon does?
You mean in your pathetic world a dual PIII is adequate?
In fact, I'm working on a dual PIII right. And no, it can't. When I need to get work done, I go to my 16 CPU cluster.
>If your application is bus limited then why didn't you
>just get an old 486 out of somebody's dumpster and just
>use that?
Do you have any clue what bus limited means? Because I'm bus-limited on a PIII, I should drop back to a tremendously slower, narrower bus? Yea, that makes sense.
Here's what bus limited means to me:
If I make one run of this sim at a time, it will finish in about 13 hours. If I start two runs at the same time, they'll finish in about 17 hours. So, I'd be better off with two single CPU PIIIs to make my runs on, then this dual PIII. This is because the memory bus, which is shared by both processors can't keep up with the memory demands of two simultaneous runs. The CPUs are data starved.
>Leave SMP concepts for those that understand it.
Considering your lack of understanding, this is actually kind of funny. A dual PIII is barely SMP anyway. He11, you're probably running 98 on it.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 05:56 PM.</EM></FONT></P>
You think a dual PIII is even vaguely comparable to a dual Alpha? And yet you call me a twit? Go find a clue, It's not even close. You can start with those SpecFP benchmarks.
Here I'll help you out:
http://www.spec.org/osg/cpu2000/results/cfp2000.html
##################################### base peak
Compaq Computer Corp AlphaServer ES40 Model 6/833 1 590 658
####### Intel Corp. Intel OR840 (1 GHz Pentium III) 1 327 335
Here are the FP rates.
This shows how well the SMP architecture scales.
########################## #CPU base peak
Compaq AlphaServer ES40 Model 6/833 1 6.84 7.63
Compaq AlphaServer ES40 Model 6/833 4 24.3 26.1
Scaling Factor 3.55 3.42
I couldn't find any Intel numbers on this page. Wonder why?
Because Intel SMP scaling sucks.
Oh, but wait ... the Alpha is only 833 MHz, it must be slower.
Yeah, right...
Sorry, it doesn't do windows though.
>Did you read the article titled "Professional Affair" in
>Toms Graphic Guides you twit?
Frankly, in my work I don't give a damn about graphics performance. Pretty pictures are for generals, the analysts look at the numbers.
>Are you saying that in your pathetic world a P3 couldn't
>do what your Athlon does?
You mean in your pathetic world a dual PIII is adequate?
In fact, I'm working on a dual PIII right. And no, it can't. When I need to get work done, I go to my 16 CPU cluster.
>If your application is bus limited then why didn't you
>just get an old 486 out of somebody's dumpster and just
>use that?
Do you have any clue what bus limited means? Because I'm bus-limited on a PIII, I should drop back to a tremendously slower, narrower bus? Yea, that makes sense.
Here's what bus limited means to me:
If I make one run of this sim at a time, it will finish in about 13 hours. If I start two runs at the same time, they'll finish in about 17 hours. So, I'd be better off with two single CPU PIIIs to make my runs on, then this dual PIII. This is because the memory bus, which is shared by both processors can't keep up with the memory demands of two simultaneous runs. The CPUs are data starved.
>Leave SMP concepts for those that understand it.
Considering your lack of understanding, this is actually kind of funny. A dual PIII is barely SMP anyway. He11, you're probably running 98 on it.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 05:56 PM.</EM></FONT></P>
"This is because the memory bus, which is shared by both processors can't keep up with the memory demands of two simultaneous runs."
You'll have to try this on an SMD P4 system eventually. I'd be interested in knowing how the dramatic bandwith increase helps out.
-Raystonn
-- The center of your digital world --
You'll have to try this on an SMD P4 system eventually. I'd be interested in knowing how the dramatic bandwith increase helps out.
-Raystonn
-- The center of your digital world --
>You'll have to try this on an SMD P4 system eventually.
>I'd be interested in knowing how the dramatic bandwith
>increase helps out.
Yea, I'm trying to line up a DDR and an RDRAM system to do some benchmarking on also. Not SMP yet of course.
In theory, there is no difference between theory and practice.
In practice, there is.
>I'd be interested in knowing how the dramatic bandwith
>increase helps out.
Yea, I'm trying to line up a DDR and an RDRAM system to do some benchmarking on also. Not SMP yet of course.
In theory, there is no difference between theory and practice.
In practice, there is.
Can't you read twit? Did I say my P3 setup compared to an Alpha setup? Well did I wanker? Those so called pretty pictures need more processing grunt than most simulations or material analysis scenarios that most Aerospace Engineers would ever come across. How do I know? Cos my friend is an Aerospace Engineer and a well renound one. Tell me what you use your 16 cpu cluster for and let me prove to the entire community what a fraud you are. Tell us about the sims you structure and the co you work for. Don't forget the final formulas as well masturbator. Check out the stats between single and dual cpu's at www.viatech.com before you make comments like "dual P3 is barely SMP anyway". Check out the major gains that are almost double speed in their SMP chipset comparison. And get off this forum you fraud. Stop wasting space. 98 doesn't support duals....even my dog knows that.
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!<P ID="edit"><FONT SIZE=-1><EM>Edited by tonestar on 04/13/01 06:28 PM.</EM></FONT></P>
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!<P ID="edit"><FONT SIZE=-1><EM>Edited by tonestar on 04/13/01 06:28 PM.</EM></FONT></P>
jollygrinch
Distinguished
- Dec 31, 2007
- 412
- 0
- 18,780
the graphics article you cite used the athlon 'a' processor. i can't speak for other people but i was under the impression that most people on this forum when saying "athlon", now, are refering to the athlon 'b' or 'c' thunderbird. <A HREF="http://www4.tomshardware.com/graphic/00q2/000515/opengl-22.html" target="_new">Test Setup</A>.
At the core of every system: "I'm sorry dave, i'm afraid i can't do that."
At the core of every system: "I'm sorry dave, i'm afraid i can't do that."
Look my point is simply this: Does it matter much weather I have an 800Mhz AMD or 800Mhz Intel unit? No, because the difference in real time in miniscule. For a real performance gain you need to step up considerably. Dual cpu's might not be for everyone but you might spend money on SCSI, RAID, a new graphics card, more ram etc. But for most power users it'll mean this. Dual cpu's, ram and graphics card. From the article you can tell that the Intel instruction set it better supported than AMD's. So in the pro graphics card arena Intel wins. You can't even consider going duals with AMD. All things considered AMD just isn't a performance option. And most consumer priced units will play games adequetly be it AMD or Intel. So therefore there is no high end AMD solution. You can't deny that.
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
>Can't you read twit? Did I say my P3 setup compared to an Alpha setup?
Sure sounded like it to me. Anybody else?
>Those sound called pretty pictures need more processing
>grunt than most simulations or material analysis scenarios
>that most Aerospace Engineers would ever come across.
>Tell me what you use your 16 cpu cluster for and let me
>prove to the entire community what a fraud you are. Tell
>us about the sims you structure and the co you work for.
>Don't forget the final formulas as well masturbator.
Ok, well one of the codes is about 500,000 lines of C/C++/Fortran. I'm afraid I can't summarize that for you in a few minutes. In fact, I can't even tell you what it does, so it doesn't really matter.
I can give you a rough, abstracted outline of the other one though. Are you familiar with the concept of an NP complete problem. Look it up. They are the hardest computational problems out there. I have two of them nested one inside the other.
The outside one is basically a Job Shop sheduling problem with up to 32 "machines". My program has to decide which of up to several hundred jobs it should give to each machine. Of course, figuring out which machine can even do each job is a big computing job in itself.
Now, once preliminary assignments are made to those machines, they have to figure out what roughly amounts to a traveling salesman problem with up to 200 "cities". Except this is orbital stuff, so everything is moving and you need high precision propagators every time you test a path. You're also doing dsp stuff for every visit as well. I say roughly a traveling salesman problem, it's actually higher order then that.
Now, while you're figuring out that TSP, you have to consider whether your initial Job Shop assignment was the best that it could be. So you need to tweak it as you go, which means you start the TSP all over again. You do this several times until you're satisfied.
By the way, all this was for just 10 to 15 minutes of simulation time. You need to simulate at least 36 hours to get a realistic run.
Oh, did I mention ... this will eventually have to run in far better then real time.
As I said, this is just a vague, abstracted description. In reality, it's kind of complicated. So go ahead and make your pretty pictures and pretend its important.
The 16 CPU cluster is just a proof of concept until we can get funding for something better.
>Check out the stats between single and dual cpu's at
>www.viatech.com before you make comments like "dual P3 is
>barely SMP anyway".
Dual P3 is barely SMP. And if you had half a clue what you were talking about, you'd know that. The guys down the hall have a 32 CPU SGI Origin that they're considering upgrading to 64. Now THAT's MP!
>Check out the major gains that are almost double speed in their SMP chipset comparison.
Sure, any MP solution will scale well if the processes fit in cache. The sim I descibed before has a fractional scaling ratio.
>And get off this forum you fraud.
LoL...
And finally, if you're right, why do you have to resort to name calling, profanity, and hearsay to prove it.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 06:54 PM.</EM></FONT></P>
Sure sounded like it to me. Anybody else?
>Those sound called pretty pictures need more processing
>grunt than most simulations or material analysis scenarios
>that most Aerospace Engineers would ever come across.
>Tell me what you use your 16 cpu cluster for and let me
>prove to the entire community what a fraud you are. Tell
>us about the sims you structure and the co you work for.
>Don't forget the final formulas as well masturbator.
Ok, well one of the codes is about 500,000 lines of C/C++/Fortran. I'm afraid I can't summarize that for you in a few minutes. In fact, I can't even tell you what it does, so it doesn't really matter.
I can give you a rough, abstracted outline of the other one though. Are you familiar with the concept of an NP complete problem. Look it up. They are the hardest computational problems out there. I have two of them nested one inside the other.
The outside one is basically a Job Shop sheduling problem with up to 32 "machines". My program has to decide which of up to several hundred jobs it should give to each machine. Of course, figuring out which machine can even do each job is a big computing job in itself.
Now, once preliminary assignments are made to those machines, they have to figure out what roughly amounts to a traveling salesman problem with up to 200 "cities". Except this is orbital stuff, so everything is moving and you need high precision propagators every time you test a path. You're also doing dsp stuff for every visit as well. I say roughly a traveling salesman problem, it's actually higher order then that.
Now, while you're figuring out that TSP, you have to consider whether your initial Job Shop assignment was the best that it could be. So you need to tweak it as you go, which means you start the TSP all over again. You do this several times until you're satisfied.
By the way, all this was for just 10 to 15 minutes of simulation time. You need to simulate at least 36 hours to get a realistic run.
Oh, did I mention ... this will eventually have to run in far better then real time.
As I said, this is just a vague, abstracted description. In reality, it's kind of complicated. So go ahead and make your pretty pictures and pretend its important.
The 16 CPU cluster is just a proof of concept until we can get funding for something better.
>Check out the stats between single and dual cpu's at
>www.viatech.com before you make comments like "dual P3 is
>barely SMP anyway".
Dual P3 is barely SMP. And if you had half a clue what you were talking about, you'd know that. The guys down the hall have a 32 CPU SGI Origin that they're considering upgrading to 64. Now THAT's MP!
>Check out the major gains that are almost double speed in their SMP chipset comparison.
Sure, any MP solution will scale well if the processes fit in cache. The sim I descibed before has a fractional scaling ratio.
>And get off this forum you fraud.
LoL...
And finally, if you're right, why do you have to resort to name calling, profanity, and hearsay to prove it.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 06:54 PM.</EM></FONT></P>
"So therefore there is no high end AMD solution."
If you define "high end" as SMP, then at this moment you are correct. Not many people need SMP though. If you're going to go with SMP, why limit yourself to only 2 processors?
Most people around here are looking for a high end workstation system. This implies a single processor system. You won't get a huge performance increase with an SMP system running most games. They simply aren't designed as multi-threaded applications. To pump up power in this arena you need the fastest single CPU, coupled with the fastest memory bus so we don't data-starve the CPU.
-Raystonn
-- The center of your digital world --
If you define "high end" as SMP, then at this moment you are correct. Not many people need SMP though. If you're going to go with SMP, why limit yourself to only 2 processors?
Most people around here are looking for a high end workstation system. This implies a single processor system. You won't get a huge performance increase with an SMP system running most games. They simply aren't designed as multi-threaded applications. To pump up power in this arena you need the fastest single CPU, coupled with the fastest memory bus so we don't data-starve the CPU.
-Raystonn
-- The center of your digital world --
>Does it matter much weather I have an 800Mhz AMD or 800Mhz
>Intel unit? No, because the difference in real time in
>miniscule.
Funny, we benchmarked a 1GHz PIII against a 1GHz TBird and found a difference of about 19%
When you're run times are measured in hours, that's kind of significant.
>So therefore there is no high end AMD solution. You can't
>deny that.
I am absolutely denying that. At least within the single CPU x86 market, AMD is THE high end solution, at least until the P4 system price becomes reasonable, and I suspect you would need to do SSE2 programming for the P4.
And I've given you a pretty good example that dual CPUs are not neccesarily the best solution for all problems.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 06:58 PM.</EM></FONT></P>
>Intel unit? No, because the difference in real time in
>miniscule.
Funny, we benchmarked a 1GHz PIII against a 1GHz TBird and found a difference of about 19%
When you're run times are measured in hours, that's kind of significant.
>So therefore there is no high end AMD solution. You can't
>deny that.
I am absolutely denying that. At least within the single CPU x86 market, AMD is THE high end solution, at least until the P4 system price becomes reasonable, and I suspect you would need to do SSE2 programming for the P4.
And I've given you a pretty good example that dual CPUs are not neccesarily the best solution for all problems.
In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 06:58 PM.</EM></FONT></P>
>If you define "high end" as SMP, then at this moment you are correct.
If I might edit your post...
"If you define "high end" as x86 SMP, then at this moment you are correct."
In theory, there is no difference between theory and practice.
In practice, there is.
If I might edit your post...
"If you define "high end" as x86 SMP, then at this moment you are correct."
In theory, there is no difference between theory and practice.
In practice, there is.
Well ergeorge you are now a certified wanker. You didn't provide any details to establish any sort of credibility, you just avoided them by frauding some job lot example. Here's why: We have 12 CNC machines at my work. 8 of them are worth more than $4 million Australian each. We also have 2 FMC (Flexible machining cells) worth millions more. We manufacture some of the most complex components for the defence and aviation industry's. All of our finite analysis is done on a single cpu machine and often compiled in less than an hour. Why do you need to simulate for 36 hours when there are only 24 hours in a day dickwad? Your still saying duallies are barely SMP, did you even go to viatech and see the benchmarks? What moron would still deny those figures? Who do you work for so I can ring them? Your probably unemployed. Hey surf the 3D ring and check out the single/dual comparisons for yourself. Check out the AMD vs Intel performance figures. When you get asked to establish some creditability come up with some real world figures not ill established stories. Check out the sites, give me the facts I asked for and until then anyone who has read the previous articles in this forum will know that your a fraud. Your not even an engineer are you?
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
Oh Ray...getting through to you is so hard. "Why limit yourself to 2 processors?" Are you serious? Go and check the cost of a quad board. "Most people around here are looking for a high end work station system" A dual cpu unit is that. If they can barely afford to go from 1 cpu to 2 cpu's there definetly not gonna be able to afford a quad setup. Check out the scene Ray and some of the prices so you can make an informed comment on this forum. Of course you won't get a huge performance increase with games...is that what having a workstation means to you fool? The word is "workstation" not "playstation". Everyone knows that games are not multithreaded. God tell us something we don't know. I never once talked about games Ray and if you had bothered to read Toms Graphics Guide those cards are designed for CAD/Cam, Animation etc. Not ONE of them is a games card. Are you on the same planet as the rest of us? See you can't deny AMD hasn't got a high end solution. Anyone who has a look a dual cpu benchmarks will know that right away (ok maybe not for playing solitaire or sending emails) but creative and productive apps for sure. Nobody but nobody can deny this. Check out www.viatech.com, 3Dlabs, Elsa, Intel, and CAD app or CAD benchmark site. For god sake don't take my word for it, read a little!!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
You are a real big man aren't ya? So far reading this thread you are the only one that seems like a fraud... It is people like you that should have the livingSHIT beat out of.
We manufacture some of the most so FUCKINWHAT you are probably a janitor at that company!
LOSER!!!LOSER!!!LOSER!!!
Have a nice day yall!
We manufacture some of the most so FUCKINWHAT you are probably a janitor at that company!
LOSER!!!LOSER!!!LOSER!!!
Have a nice day yall!
MJDUNN? Are you some kinda of crack whore? Have you actually got anything to say? And no I'm a mechanical engineer not a janitor. Why did you even make a post? Have you got any real facts to offer? Come on bitch, show us you've got a brain or go back to the gay porn site you came from. Post something usefull or at least tell us about the dozens of guys your mum srews every week. Trash head.
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
"Cock-a-doodle-do" is what I say to my girl when I wake her UP in the morning!!
TRENDING THREADS
-
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Facebook
Twitter
Instagram