Tech specs: What are the Core 2 Duo power equivalents?

[MrSpkr]

Okay, so these are the recommended stats:

Pentium IV 3.2 GHz or equivalent speed AMD processor or better

Video Card: GeForce 6800 or Radeon x850

(256 Megabyte VRAM or better)

Sound Card: DirectX 9 compatible Sound Card

System Memory: 1 Gigabyte or more RAM

Hard Drive Space: 1 Gigabyte

What is the power equivalency for a Core 2 Duo processor? Does a Core 2 Duo processor running at 1.6 GHz work as well as a P IV 3.2 GHz processor? How about a Core 2 Duo running at 2.0 GHz?

Need to know because I am purchasing a new laptop in the next couple of weeks, and would like to get one that runs CMSF at a decent pace with lots of the pretty graphics features.

Thanks.

Steve

[Pzman]

A Core Duo (pre Core 2 Duo) at 1.67Ghz is faster than a 3.8Ghz P4; I think its safe to say that if you have a C2D, 1.5Ghz or faster the game should run fine. People really need to learn that higher Ghz does not equal better performance, especially now with the new line of Dual Core CPUs. Even the Core Solo chips (Core Duos with one core disabled) are faster than the old 3.8Ghz P4s.

[Redwolf]

Reposting my post from the general forum:

Getting fed up with the misinformation in the other threads here, so

here's it in short. Different processor architectures have different

performance per MHz.

Basically you have to tell apart these three lines:

</font>

• Intel Netburst (Pentium-4, P4M, P4-based Celerons)</font>
• AMD64s (Athlon 64, Athlon FX)</font>
• Newer P3-derived Intels (Pentium-M, Core Duo, Core2)</font>

Intel Netburst/P4 performed well only for very specific applications,
mostly video encoding and some selected graphics tasks. Gaming was
bad, as was power consumption,

Compared to a AMD64, a P4 needs much higher clockspeed for the same
performance. A 2.2 GHz AMD64 plays about as well as 2.8 GHz
Netburst/P4.

Having realized their error with the lame and power-hungry Netburst
CPUs, Intel developed the Pentium-III line into the Pentium-M, Core
Duo and Core2 chips. These compare as well or even better than AMD64
at the same clocks.

Here's an approximate table of performance per clock for gaming, with
AMD64s placed at 100:
</font>

• 100% AMD64s, Athlon 64/FX, Opteron - 100%</font>
• 70% old Pentium-4s with less than 800 MHz FSB</font>
• 80% Pentium-4s with Northwood core and 800 MHz FSB</font>
• 78% Pentium-4s with Prescott core</font>
• 105% Pentium-M</font>
• 110-115% Core Duo</font>
• 120-130% Core 2 Duo/Quad</font>

[update]

The cache size plays only a negligible role in the AMD64 world, don't

be fooled by the AMD64s with 512 or 1024 KB cache. It only brings a

couple percent at best. Going below 512 KB isn't recommended, though, it quickly brings you down 10-15% in some tasks.

In the Core2 world, however, cache size is much more valuable. The 512 KB cache Core2 based Celerons perform between equally fast to Conroes for e.g. some video/audio encoding, but are 30% slower for tasks like C++ compilation.

[ July 18, 2007, 12:14 PM: Message edited by: Redwolf ]

[Jim Cobb]

Quick question: I have an 2.07 gigahertz AMD Athlon XP.

Will CM:SF run on this?

Thanks.

Jim Cobb

[Battlefront.com]

Hi Jim,

Yeah, it should. Charles coded the game on something slower than that IIRC Obviously the more speed the better, but that is always the case. If you play WeGo you have it easier than if you want to play RealTime simply because framerate doesn't affect WeGo playability as much as in RT.

Redwolf is very correct. For two decades now computer companies would have you believe that chip speed is the most important factor. Never has been! The Mac's Motorola (then later Motorola/IBM) chips, for example, always performed better than Intel or AMD MHz for MHz. Part of that was because the processor was always more advanced (32 vs. 16 bit, then 64 vs. 32), but also partly because Apple put in quality subsystems where most PCs put in crap to keep their prices down. Which really hurt Mac sales because people thought the cheaper PCs were faster because they had higher clock speeds, when in reality they were the same if not slower.

I remember one time seeing some Pentium II computers compared against each other. There was a less expensive, slower P2 system that outperformed a name brand more expensive one with a faster P2. The reason? IIRC it was the backside catch and the databus that were used.

This was a trick of Pacard Bell. If the fastest procesor was 120MHz (remember those days?) that's what they would have in their computers. Then to keep costs down they wouldn't have a cache, no math coprocessor, etc. Joe and Jane Consumer just went and looked at the MHz and that that's all there was too it. And they became victims of the evil plan

Steve

[ July 18, 2007, 02:31 PM: Message edited by: Battlefront.com ]

[aka_tom_w]

And RAM Lots of RAM

512 megs of Ram is the absolute Minimuum

1 or 2 Gigs of RAM will make a world of difference to your RealTime fun FPS action.

I would like to speculate the ANY decently fast CPU that was brand new within the past two years will be fast enough to play RealTime if you run with 2 Gigs of RAM and any decent (new in the last two years) video card.

Bottom line:

Buy More RAM

(you can also just throw in a FASTER harddrive while your at it and you WILL see and feel a noticable speed improvement in your computer and gaming FUN)

[dan/california]

For approximately $15,000 Apple will sell you a system with 16 gig of ram and enough of everything else to require its own powerline. How big a map would that run? :eek:

[Redwolf]

Originally posted by dan/california:

For approximately $15,000 Apple will sell you a system with 16 gig of ram and enough of everything else to require its own powerline. How big a map would that run? :eek:

I don't think that RAM will be the factor limiting map size. A larger map means more numerous and more expensive LOS checks, and that is more CPU-driven than memory driven, and the CPU suffers more than memory as map size increases. So you get to a CPU crawl earlier than you run out of RAM.

Any socket 771 or Socket F system will give you 16 GB easily, you don't have to spend $15,000 on it

2x 1 GB DDR2 sticks can be had around $60 now. Anybody with a DDR2 capable system should go to 2 GB before installing any current game, IMHO, it's the price of a game.

[gibsonm]

Well I can run a pretty big map already and I have an “old” Intel Core Duo box.

[Battlefront.com]

The problem is that the bigger the maps get the more RAM the system needs to store them. When they aren't kept in RAM things get slooooooow. Too much swapping and what not. Or so I've been told

But Redwolf's comments are still pretty acurate. If you don't have a good card or CPU it won't really matter too much if you have 16GB of RAM (holy crap that's huge!). On the other hand, if you have a great card and CPU, but paltry RAM, you'll get bottlenecked when the scenarios are big.

Steve

[Battlefront.com]

For me I have a 2.33 GHz Core Duo with 2GB of RAM and 4MB Cache. The thing that hurts me is a fairly so-so ATI card. But it's an Apple Mac Book Pro, so I ain't complaining Overall I'm very happy with the performance on my system.

Steve

[GriffinCheng+]

One moment, if I am to upgrade my old rig from a 2.8GHz P4-C, does that mean C2D would be a better choice over AMD?

I can barely play ToW so this is quite essential to me.

TIa.

[Cameroon]

If you're looking for the best CPU performance, the C2D is easily the reigning champion.

[Edit]

I thought about looking up all the various benchmarks, but they all say the same thing. Just check out any review of the C2D from any of the reputable review sites. C2D wins across the board basically.

[Redwolf]

Originally posted by GriffinCheng+:

One moment, if I am to upgrade my old rig from a 2.8GHz P4-C, does that mean C2D would be a better choice over AMD?

I can barely play ToW so this is quite essential to me.

Not necessarily, because there is a price war going on and the AMD CPUs might be cheaper for the money at the same performance. You take a price list and add 20% clockspeed to the AMD64s, then compare prices.

Also, mainboards for AMD CPUs are generally cheaper. But on the other hand mainboards for Intel CPUs with Intel mainboard chipsets have better driver support and are generally a little more solid than that NVidia chipset (not graphics) stuff you find in most better AMD-CPU systems.

[Wisbech_lad]

PS, long time no see MrSpkr! Welcome back, you paleoconservative lawyer you!

(and no, I don't count the Peng thread as being back...)

[MrSpkr]

Originally posted by Redwolf:

</font><blockquote>quote:</font><hr />Originally posted by GriffinCheng+:

One moment, if I am to upgrade my old rig from a 2.8GHz P4-C, does that mean C2D would be a better choice over AMD?

I can barely play ToW so this is quite essential to me.

Not necessarily, because there is a price war going on and the AMD CPUs might be cheaper for the money at the same performance. You take a price list and add 20% clockspeed to the AMD64s, then compare prices.</font>

[MrSpkr]

Originally posted by Wisbech_lad:

PS, long time no see MrSpkr! Welcome back, you paleoconservative lawyer you!

(and no, I don't count the Peng thread as being back...)

Thanks! Been a busy two years -- have recently opened my own practice, just so I can have more time to berate Joe Shaw and company.

Steve

[Jim Cobb]

Thanks, guys. I have a Gig of Ram. The trouble I'm having a cursor lag/jerkiness.

t

[Redwolf]

Originally posted by MrSpkr:

</font><blockquote>quote:</font><hr />Originally posted by Redwolf:

[Hmm. So you are saying that a 1.0 GHz C2D runs at the same speed as a 1.2 GHZ AMDx2?

Interesting.

Here is what I am looking at:

Will this rig play well when using high graphic features and large maps?

</font>

There are no 1 GHz Core2s but anyway. Yes, in general a 2.0 GHz Core2 is about as fast as a 2.4 GHz AMD 64. However, the 20% advantage is a minimum, I have seen Core2 pull 30%, even 50% ahead. But in general you should calculate with 20-25%.

You should take a card with more memory bandwidth if you want high resolutions with AA/AF and games like CM:SF, which I assume has high polygon count (lot of vehicles and no LOD) and is pretty light on shaders (no face animation etc.).

[Vern_S]

Here is a website that compares all cpu's

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

[panzermartin]

According to the test I got a 711 CPU mark, guess that is ok for CMSF with 1GB ram and a 7600gt?

[Dirtweasle]

Thanks Vern!

[Vern_S]

No Problem...

Basically if your CPU scores greater then 500 and you have a decent video card, 6800 and up, you are set for good gaming!

[__Yossarian0815[jby]]

Originally posted by Vern_S:

Here is a website that compares all cpu's

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

dual cores seem to be severely overrated in that benchmark. the dual core variants have more or less double the scores that their single core variants have. In the real world that isn´t true (yet)

I would check out actual gaming benchmarks to get an idea what you want.

[Redwolf]

Yeah, that benchmark is useless. First of all it doesn't explain exactly what it is doing, and second it doesn't even try to evaluate per-core performance (which is all that counts for CM and ToW).