T6600 vs P8600

Is there much of a real world difference in the T6600 and the P8600? I know the technical differences (2.4 vs 2.2 GHz, 3mb vs 2mb cache, 35w vs 25w), but does this amount to much of a difference? I'm looking at purchasing a Dell and the difference in price is nearly $200.

Most of my uses would be general web and office stuff, with some occasional light gaming.

Any input would be appreciated, especially if someone has experience with both.

 

According to passmark the p8600 is 6.5% more powerful than the t6600

But the p8600 has a TDP of 25watts compared to the 35watts of the t6600 and supports visualization.

http://ark.intel.com/Compare.aspx?ids=35568,37255,

http://www.cpubenchmark.net/cpu_list.php

 

Don't worry about TDP; the T6600 will use less power than the P8600. Unless you know need virtualization, you really don't need the P8600. $200 for less than a 10% increase in clock speed isn't worth it.

 

LOL what nonsence is this? If you don't know what your talking about , don't write... The P8600 is much better than a T6600... it uses less power , has higher clocks (2.4 GHz vs 2.2 GHz) higher L2 cache and higher FSB.... so it is a better choice... it will help in increasing battery life so P8600 is better than T6600... as usual trottel spreads misinformation...

 

Can you please elaborate on the "much better" part? I know spec wise it's better, but is there going to be a noticeable difference? As the first reply mentioned, I don't consider 6.5% much of an increase in performance for an 24% increase in laptop price.

 

You should follow your own advice.

1. Based on the laws of physics, the T6600 will consume less power than the P8600.
2. A 9% increase in clock speed nets less than that much more performance. Higher FSB and 50% more L2 cache don't do much.
3. The price difference of $200 is not worth it if he doesn't need VT and he only uses the processor for the tasks he listed.

 

About TDP.
It is the maximum the cooling system needs to draw to cool down the components. At idle, both are about the same consumption.

True that the P8600 is better specs wise, but unless you plan to stress the CPU, it is not worth it. The T6600 is quite capable for regular users.

Would you mind telling us your usages? Are you planning on renders? Database handling? Numbers crunching? Calculations?

Thanks!

 

Listen to Trottel guys. Even if the OP does intensive calculations, $200 is still far too much for only a modest increase in speed. TDP has already been shown by users with both P and T series (like K-TRON) to be an unreliable measure of power consumption. The only factor important factor is if VT is needed.

 

I don't even know what VT is. I did a search and from what I read it's not relevant to anything I'll do, so I think it's a nonfactor.

The most intensive use of the computer would be for gaming, which won't even be that often as I have a desktop for that. I do some light 2D AutoCAD and photoshop work, but that's really not intensive.

 

Then the T6600 satisfies your needs with ease.

 

i just had 2 laptops with these exact same processors. the p8600 isn't worth the money. the t6600 runs cooler because most of the time if you're just surfing the web or what not the cpu isn't stressed and the cpus will run at a lower clock with intel speedstep. the p8600 will run at a minimum of 1.6ghz and the t6600 will run at a min of 1.2ghz.

 

The T6600 falls in the 35 watt TDP, but it doesnt use 35 watts
the P8600 falls in the 25 watt TDP, but that doesnt mean it uses 25 watts.

I would expect the P8600 to use 23watts or so under load, and the T6600 to use close to 27watts.

Intel is smart with there marketing, processors which do not fall in the less than 25 watt TDP zone get bumped into the 35 watt TDP bracket.
So a 26 watt processor will fall in the 35 watt TDP bracket.

Intel doesnt label the processor as 27Watt, cause than people would see than a 2 watt difference isnt worth the extra money for the P series chips. So they keep a 10 watt difference, which is enough for people to pay more for the P series chip.

In reality, they will use close to the same power, and run the same temperatures. You can thank Intel's Binning methods for this

K-TRON

 

K-TRON, is it ok if if I ask you about your tests with P and T series?
That could be helpful for an Intel Guide I wrote a week ago, so I can add that as a testimonial and help people.

If you dont mind, and you fully agree with it. PM if you want. Thanks!

 

Sure thing
I tried an indentical clocked T9600 and P9600 series chip in the same laptop - HP DV5

The real life differences I saw under full load between the T9600 and P9600 was 3C under full load. The same heatsink, thermal material, and fan were used.
What we were talking about in the other thread is Intel's binning processo

Say Intel tests one processor out of say 1000 chips made per silicon wafer. If that processor uses 24.99 watts, it fall in the 25 watt TDP.
If that processor uses 25.01 watts, it fall in the 35 watt TDP
Intel doesnt test every processor, so in say that batch of 1000 processors, you are getting half of them to use 24.99W, and half using 25.01 watts. Thus the difference in power consumption is negligable.
If you want to believe that the T series came out using 35 watts, and than a week later Intel releases 25 watt chips, you can. A 40 % power decrease doesnt happen overnight.


K-TRON

 

Thanks a lot K-TRON for the info.
Can I post that in my Guide?
Would you like to be quoted/referred to, or stay anonymous?

That sure comes in handy, and should help me improve the Guide with some more accurate facts.

 

K-TRON, thank you for your post. It's good to know how Intel bins it's chips, and it does make sense. I had this impression that the P8600 was a must have, and worlds better than the T6600, but you helped me decide the T6600 was more than adequate for my needs.

 

K-Tron/Serg
Could it be possible the "P''s are just a "T" with their highest voltage set a bit lower which would give them the ability to sell it as lower TDP ?
P = max 1.1**V
T = max 1.2**V
Probably enough of us here to work that out.

 

I really want to do a processor comparison soon. I have 3 T4200's, 2 P8600's, a T6600, a pin modded celeron 900 at 2.93Ghz and a stock celeron 900 at 2.2Ghz I can use. I would do all the testing on the exact same laptop under the exact same conditions to test for power consumptions as reported by rm clock both at idle and using orthos. Also maybe with how long the battery lasts under orthos and at idle, but that might take forever, even with a 4 cell battery. I have a test on wednesday, a paper due on friday, and I need to go to a concert and write a paper about that by the friday after next, as well as redoing half the wiring in my car, so it might not be for a few weeks that I can get it all done and prepared for here.

 

Trottel, if you do the test, PM me.
And if you allow me, and want to, I will post the results in my Guide, with credits to you of course, and a link in the end to your thread if you do one.
That info will come in handy, eliminating all variables but the CPU.

Thanks!

 

Of course, I would be happy to to.

 

GRRRRRRRRR I CANT POST!!!!
Thanks a lot!
Keep me up-to-date. That info is priceless!

Oh and to the OP, the P series use less Wattage, in theory, that is why they are under the 25w bracket.
Third attemp

 

It's not theory, it's based on testing a sample from a batch of CPUs.

 

Sorry, that post came out wrong.
I wrote it three times, and by the last time I was already annoyed. Let me redo it.

The P series CPU uses less Watts than a T series one, but it is not necessarily the 40% difference quoted by them. A T will perform as good as a P and the other way around too. Both, unless stressed, will consume more or less the same, and for 125 pounds it is not worth it.

I think that came out better, no?

 

Isn't TDP = Less watts to cool it - which i think i learned from 1 of your posts somewhere Serg

I'm thinking they do a run of processors setting the max voltage @ 1.2*** V and name them T**** then a differant run setting the max voltage @ 1.1***V and name them P****
If we set up a "Post your original Penryn CPUID voltage thread " i think we might see that.
Less volts = less heat which = lower TDP Guaranteed.

My thinking anyway - mostly learned from you guys

From Intels info....... http://www.intel.com/products/processor_number/about/core.htm
"T" = Mobile highly energy efficient processors with TDP 30-39W
"P" = Mobile highly energy efficient processor with TDP 20-29 W

Obviously they then get branded 25W or 35W depending what side of 30 they fall in.(which they'll know if their setting the voltages lower/higher)

 

So I have a question regarding idle power consumption for the T and P series processors. Given that Intel's SpeedStep depends on Front Side Bus speeds and scales back to the lowest multiplier of x6.0, would a laptop with a T series processor use less power when idle since it has a FSB speed of 800 MHz and therefore idles at a lower clock speed of 1.2 GHz (200 MHz x 6.0) vs. a P series processor which has a 1066 MHz FSB and therefore idles at 1.6 GHz (266 MHz x 6.0 )?

 

Probably would.

 

my 2 lowest multipliers on P7450 (1066 FSB)
6x Super LFM....800Mhz
6x Normal........1.6Ghz

Generally it will idle @ 1.6Ghz
I'm not sure but maybe it only uses 6x superLFM in sleep/hibernate.

T9550 and above look like they are all 1066 bus, below that looks like they are a mix.

Can't lower 6x voltage below 0.925V as its locked - idle temp is ~37/43

 

I don't think a single one of these posters even has a P8600. Let me tell you, my P8600 washes my dishes AND lends a listening ear when I need to get something off my chest. Well worth every penny of your $200.

 

Wow, do you even have a T6600? I have both.

 

The T6400/T6500/T6600(since they're essentially the same thing) are enough for most of widespread computing. Even most games and applications run fine on that. People tend to overestimate their needs or confuse them with what they want, hence why companies can make money off upgrades like that 200$ one.

 

I agree there.
That is why the i7 820 is $350 more! hahahahaahah just for a slight 10% increase overall performance (perhaps up to 15% but I dont think so).

And the RAM upgrades are highway robbery! About 100 for a 2x2GB RAM? hahahaha funny, even more when it is DDR2, that you get for 40 or so

 

I just upgraded my T6400 with a P8600 and the computer runs cooler and you can tell it's a bit quicker when opening certain programs.
Go for the P8600 mate

 

I agree ... get the P8600... its more powerfull.. newer tech.. higher speed FSB and L2 cache.. and not to mention cooler and can be undervolted more so your battery life can increase.. T series run much more hotter and if i had a chance i would get a P series processor too..

 

Check out the link in my signature about Intel CPU's, go to the T and P chapter, at the end, there is the proof, confirmed by K-TRON that there is little difference in both.

For over 100 for a .2GHz speed, is not worth it if you ask me.

 

Whilte I don't disagree with either you or K-tron, Serg, I'd just like to point out that K-tron's logic concerning Intel's "branding" of CPUs is more or less in the "your mileage may vary" category. So long as a CPU crosses 25W TDP, it goes to say the 35W category, but in that category, you could have individual CPUs which use from 26W to 35W themselves.

So while K-tron's T series might've used little over 27W, maybe another one, for x or y reason, could use up to 30W.

It's the same logic that some CPUs undervolt better than others(despite being the same model) due to differences in voltages and material resistances of a given batch.

 

Yes, but at the same time you can have a 35W rated CPU and it consume 20W. They're all from the same place, so just b/c testing the first CPU of a batch has a TDP of 26W (so it would be rated at a 35W) doesn't mean all the CPUs in that batch will be closer to 35W than 25W. So, it's possible a T can consume less power than a P. It's all marketing to think that P is always better, since on average, the difference between a P and T processor is negligible.

 

Well yeah, that's why I'd say these things fall under the "your mileage may vary" category as each CPU, depending on external factors during when it was built, can alter.

I'd say the TDP is negligible at best since even assuming it was a 10W difference, that would be at max load, which we all know maybe 80% of users rarely if ever reach.

 

I don't wish i disrespect what K-Trons saying either but as that info is going in a guide maybe you need to look at this.

from the Intel site the T and P TDP crossover appears to be 30
From Intel
T = Mobile highly energy efficient processors with TDP 30-39W
P = Mobile highly energy efficient processor with TDP 20-29 W
http://www.intel.com/products/processor_number/about/core.htm

Also if we ran a thread looking at voltages i think we'll find the P series have there max voltages set lower - Factory undervolted - thereby making them a sub 30 TDP...meaning they get a 25 TDP "P" spec.

I agree with K-Trons "mileage can vary" theory but remember Intel must also have some degree on quality control to push out the door something labeled P versus something labeled T.
By just making the voltage 1.1**V for a "P" versus 1.2**V on a "T" is how i think they achieve that.
The Intel Spec sheets don't show that but i think that's what we'd find.

My Stock max V on this P7450 was 1.138V

 

This, my friend, is called marketing.
Indeed P may consume less, but again, when idling and normal usage it is negligible, a slight difference on them. When both are stressed, yes you will notice a difference, not highly noticeable, but there is...

For regular usages, the P and the T are the same...basically, if at same clock or small difference.

ANd yes, you can get a T that consumes either 26 or 30 or 32, yeah, it is a range, but that is when stressed, it would be insane if it would consume the max all the time...why buy it then?

 

Yes i understand that Serg - from learning here,
What i was getting at was the actual spec differance not the real world differance.
If Intel set the same clock "T" processor to a lower max voltage we now have a "P" is what i'm getting at.

I don't think they have the same max voltage verus the theory that their all stock "T"'s but if they fall into below 25 TDP they become a "P"

 

Is it possible for intel to adjust a cpu`s clock speed and features cache size and vt and others after the processor is made.

Basically 1 cpu then enable and disable features depending on the price.

If i remember with amd, you could join two laser cut tracks on the out side of the cpu to change the clock speed with a pencil

 

yes.

Yeah I remember that, it unlocked the multi so you could change it in the BIOS.

 

The laser cutting stayed for a long time. At first you could unlock it by bridging the connections on the old Athlon Thunderbirds and maybe some Palominos I think. Then AMD simply had their processors with unlocked multipliers. Then they locked the multipliers sometime in 2003, and bridging the connection would not unlock it.

My Applebred Duron had an unlocked multiplier, but to my dismay my two subsequent Barton 2500+'s were locked, one of which was manufactured right after the cutoff date.