well let's see if the P4 can preform in the future... it might be king if Intel can ramp clockspeeds really fast...for a while anyway, until it gets "Hammered" down 
hahaha, Intel, hahaha IM LAUGHING, my 1GHz Tbird and A7V stomps a Pentium4 in the mud and spits on it...
Brotherman I also appologize I thought that you were directing it towards me
I really think that the next year is going to be really interesting to see computer wise. The processor war is really heating up and we the user are who benefits from this war. Better prices and better processors
I really think that the next year is going to be really interesting to see computer wise. The processor war is really heating up and we the user are who benefits from this war. Better prices and better processors
NOS there you go again speculating that the P4 will stomp the athlon, just as you did the night before... well it didnt "Stomp" it and it wont. I admit, the p4 design is aimed at the future and when more and more software gets optimized for it, it will perform better. Until then though, a clock v. clock comparison shows that an athlon beats it in most tests... even when the p4 has a 300MHz lead, the athlon can keep up in most cases (except q3a and mpeg encoding/decoding). By the time SSE2 builds up support, AMD may be at its Athlon 2/3 (clawhammer) which has SSE2 support as well. Also, this chip is touted to start around 2ghz (maybe even faster) and should be able to compete in MHz wars as well as IPC wars. Either way, i think the p4 is a good chip for "some" people, the same way the Athlon is good for others. Also, its good fot the consumer to realize that they can now get a >1ghz cpu for ~$300.
life is good... face it, intel will never be what it once was, there is plenty of room in the market for both anyway.
alin
life is good... face it, intel will never be what it once was, there is plenty of room in the market for both anyway.
alin
ragiepew time will tell won't it. But sse2 support for AMD is at leat a year away and just like them to not support there own products but let intel do it for them. LOL!!!!!!!!
NOS... ehh, AMD is doing something smart by adding SSE2 support. They do not have the money to advertise and support 3dnow! as much as they should, and ill admit... intel is doing a better job w/ SSE than AMD w/ 3dnow... anyway, by picking up SSE2 they have that support plus the support of 3dnow!. You are right, time will tell, and yes a year away is a long time... especially in this industry. However, i do not think SSE2 will garner "that" much support for about another year. We also have palomino (dong want to speculate on its performance) and also AMD stated it would raise the ev6 bus to 333 if needed (when 166 DDRSDRAM is available). the ev6 bus is capable of going to 400mhz although when paired w/ DDR it still wont provide the moster bandwith that dual channel RDRAM provides, although it will help in other areas. Anyway, we could argue back and forth forever on this... well just have to wait and see, but for now... Athlon didnt get "steamrolled" like you said it would .
.i kept hearing about how amd is going to have ddr sdram? and intel going to have RDAM? what are the differences?
RDRAM is a proprietary format. It transmits at 800MHz on a very narrow 16-bit bus. This means that it is easier to put multiple channels on a system.
200MHz DDR (400MHz) RAM would have the same bandwidth as a Dual-Channel PC800 configuration and have noticably lower latency if it could be implemented properly. Along with a 400MHz EV6 bus protocol it would be a sweet setup.
PC2100 DDR (on the AMD760 for example) would transmit at effectively 266MHz with a 64-bit data width, giving it slightly lower throughput than the dual-channel DRDRAM but it does not need to be installed in pairs and it has a lower latency and slightly lower power consumption. Micron is working on PC2600 right now (333Mhz/166DDR) so we'll see...
Perhaps a server setup with a Dual CPU Athlon and a dual channel PC2100 DDR. That would be sweet and with 4DIMM slots per channel, it would be an easy and inexpensive way to allow for massive amounts of memory.
P4 still relies on the old GTL signaling protocol (albiet greatly enhanced) and while its a little easier to implement, the effect is that the Athlon will still have a greater bus throughput if it's EV6 is pushed to the max. Especially in dual CPU setups- Quad EV6 would whoop up on a Quad P4 bus. Sadly, I don't now if that will happen ever.
I like this recent trend of pushing bus speeds and memory speeds up instead of releasing faster CPUs. Its something that's long overdue. We had only grown our bus speeds less than 3x since the 486 processor (fastest 486 bus was 50MHz) while CPU speed has increased by probably about 100 times (first 486 chips were 16MHz and took 6 cycles to do some math).
Now that we've triple the bus speed again we have a little headroom to work with. Don't expect Intel to be happy with 400MHz. I suspect the move to the new P4 socket might go along with the introduction of PC900 or PC1000 DRDRAM or one of the QDR technologies. Maybe not- but its a possibility.
I wonder what bus AMD will use for Hammer??
Maybe they'll develop EV6+? *grin* 600MHz EV6.
QDR SDRAM is in the works too. That would go nicely with a 400MHz EV6 bus. 200MHz*4=800MHz.
Rambus would have to kick their memory up to 1.6GHz if they want to use the same format and keep up with 200MHzQDR memory. Imagine the pricetag on a few hundred MBs of that stuff!!!
*waiting to see the future*
Eric
200MHz DDR (400MHz) RAM would have the same bandwidth as a Dual-Channel PC800 configuration and have noticably lower latency if it could be implemented properly. Along with a 400MHz EV6 bus protocol it would be a sweet setup.
PC2100 DDR (on the AMD760 for example) would transmit at effectively 266MHz with a 64-bit data width, giving it slightly lower throughput than the dual-channel DRDRAM but it does not need to be installed in pairs and it has a lower latency and slightly lower power consumption. Micron is working on PC2600 right now (333Mhz/166DDR) so we'll see...
Perhaps a server setup with a Dual CPU Athlon and a dual channel PC2100 DDR. That would be sweet and with 4DIMM slots per channel, it would be an easy and inexpensive way to allow for massive amounts of memory.
P4 still relies on the old GTL signaling protocol (albiet greatly enhanced) and while its a little easier to implement, the effect is that the Athlon will still have a greater bus throughput if it's EV6 is pushed to the max. Especially in dual CPU setups- Quad EV6 would whoop up on a Quad P4 bus. Sadly, I don't now if that will happen ever.
I like this recent trend of pushing bus speeds and memory speeds up instead of releasing faster CPUs. Its something that's long overdue. We had only grown our bus speeds less than 3x since the 486 processor (fastest 486 bus was 50MHz) while CPU speed has increased by probably about 100 times (first 486 chips were 16MHz and took 6 cycles to do some math).
Now that we've triple the bus speed again we have a little headroom to work with. Don't expect Intel to be happy with 400MHz. I suspect the move to the new P4 socket might go along with the introduction of PC900 or PC1000 DRDRAM or one of the QDR technologies. Maybe not- but its a possibility.
I wonder what bus AMD will use for Hammer??
Maybe they'll develop EV6+? *grin* 600MHz EV6.
QDR SDRAM is in the works too. That would go nicely with a 400MHz EV6 bus. 200MHz*4=800MHz.
Rambus would have to kick their memory up to 1.6GHz if they want to use the same format and keep up with 200MHzQDR memory. Imagine the pricetag on a few hundred MBs of that stuff!!!
*waiting to see the future*
Eric
"I wonder what bus AMD will use for Hammer?? "
The Hammer will most likely be using AMD new LDT bus
"The Hammer will most likely be using AMD new LDT bus
-------------------------
Whatever it is its has to be FAST and STABLE
that says it all...
try:
Whatever it is IT has to be FAST and STABLE
try that...
honestly tho, i've never seen a sig be wrong for such a long period of time...
Whatever it is its has to be FAST and STABLE
that says it all...
try:
Whatever it is IT has to be FAST and STABLE
try that...
honestly tho, i've never seen a sig be wrong for such a long period of time...
Toysrme its funny that no one else including myself ever caught it. LOL I think I will leave it that way anyway's
the truth.. I know where it is.. I filed for patent no. 290076356645453342666-11 in 1991 on "Truth" .. so everyone pay me royalties or you're liars
Blah blah... buy/use what you like. If you are satisfied with what you have then thats all you need.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter