Athlon 64 Revision E: Unofficial DDR500 Support
by Anand Lal Shimpi on July 11, 2005 12:05 AM EST- Posted in
- CPUs
The New Memory Speeds
There are a total of four new memory dividers unofficially supported by the Rev E Athlon 64s, but not all of them can be used by everyone. In order to understand why, you have to understand a bit about how memory speed is calculated by the Athlon 64's on-die memory controller.In a Pentium 4 system, the memory controller is located on the chipset, and derives its clock from the FSB frequency of the CPU through the use of a FSB:DRAM clock ratio. For example, with a 1:1 clock ratio, a 200MHz FSB clock would result in a 200MHz DRAM clock.
The Athlon 64 is a bit different, since it does not have a conventional FSB. So, instead of the memory clock being determined by a ratio of the FSB clock, it is determined by a few factors.
The basic equation is this:
DRAM Clock = CPU Clock / (ceil(CPU Clock Multiplier/Memory Divider))
Most of the elements of the equation are pretty obvious; the DRAM clock is the resulting memory frequency. Note that this is your non-DDR memory frequency. For example, if the DRAM Clock is 200MHz, we're talking about DDR400; if it is 166MHz, then we're talking about DDR333.The CPU Clock is the final CPU clock of your processor, which is made up of two components: the Hyper Transport clock and your currently selected clock multiplier. The HT clock is 200MHz by default, but can obviously be overclcoked. The CPU Clock Multiplier is set at the factory, but lower multipliers are unlocked for Athlon 64s, while all multipliers are unlocked for FX processors.
The Memory Divider is a ratio supported by the CPU's memory controller, and it is this set of ratios that has been expanded in the Rev E memory controller.
Finally, there's this "ceil()" function. The ceil() function is a pretty basic mathematical function that returns the smallest integer value greater than or equal to its argument (the number passed to the function in the parentheses). For example, ceil(5.5) = 6, and ceil(10.1) = 11. Pretty simple, right?
So, you plug in all of the variables of that equation, and solve, and you get your final DRAM clock.
You'll notice one very important thing about this equation: the DRAM clock is dependent on the Athlon 64's clock speed , which means that in order to achieve the same memory speed on all processors with differing clock speeds, the memory divider is going to have to, well, vary.
Prior to the Rev E CPUs, the Athlon 64's memory controller supported enough dividers to allow for DDR400 to be supported at all clock speeds; from 1.8GHz all the way up to the present-peak of 2.8GHz on the Athlon 64 FX-57. The Rev E CPUs support those same dividers, but add the following:
13/12, 7/6, 5/4 and 4/3
If you plug these ratios into the equation above, you can come up with a list of the new memory speeds unofficially supported by Rev E CPUs:| CPU | Clock Speed | Memory Dividers | |||
| 13/12 | 7/6 | 5/4 | 4/3 | ||
| AMD Athlon 64 FX-57 | 2.8GHz | 215MHz | 233MHz | 233MHz | 255MHz |
| AMD Athlon 64 3800+ | 2.4GHz | 200MHz | 218MHz | 240MHz | 266MHz |
| AMD Athlon 64 3500+ | 2.2GHz | 200MHz | 220MHz | 244MHz | 244MHz |
| AMD Athlon 64 3200+ | 2.0GHz | 200MHz | 222MHz | 250MHz | 250MHz |
| AMD Athlon 64 X2 4800/4600+ | 2.4GHz | 200MHz | 218MHz | 240MHz | 266MHz |
| AMD Athlon 64 4200/4400+ | 2.2GHz | 200MHz | 220MHz | 244MHz | 244MHz |
The above table is comprised of all of the Socket-939 Venice (90nm Rev E) cores currently on the market, as well as the Athlon 64 X2 processors, which also support the new dividers.
You'll notice that not all of the dividers are useful, some resulting in the same old 200MHz DDR400 memory clocks while others offering duplicate speeds (e.g. the 7/6 and 5/4 dividers with the FX-57).
But at the same time, a number of them produce some very interesting, and potentially useful memory configurations without ever overclocking your CPU or the Hyper Transport bus. For example, at 233MHz, the Athlon 64 FX-57 can now run with unofficial DDR466 memory. And at 250MHz, the Athlon 64 3200+ can use DDR500 memory.
At DDR466, you get approximately 15% more memory bandwidth over a standard dual channel DDR400 configuration with an Athlon 64. At DDR500, you get a full 25% increase in memory bandwidth.
Historically, the Athlon 64 hasn't really been memory bandwidth bound, since the move to Socket-939, which gave it a full 128-bit wide memory bus, and more bandwidth than these CPUs could use.
With the move to dual core however, the effective memory bandwidth that each core gets is significantly reduced, as they both have to share the same 128-bit wide memory interface normally dedicated to a single processor. So in theory, the new dual core X2 line of processors could be a good candidate for these new memory dividers.
The other situation where higher clocked memory is important is with higher clock speed CPUs. The faster that your CPU clock gets, the quicker it can process data and thus, the faster that it needs information and the more memory bandwidth that it needs.
The lower clocked CPUs are less likely to see any real performance difference, with DDR400 being more than sufficient for their needs.
Facebook
Twitter
RSS
37 Comments
View All Comments
Murthunder - Wednesday, July 13, 2005 - link
So what is the best memory for good overclocks yet is still stable? I originally tried a pair of Corsair TWINX1024-4000 Pro XMS sticks and my DFI LANParty UT nForce4 Ultra-D board would fail to post. I swapped the XMS for two 512MB sticks of Kingston ValueRam DDR333 c2.5 and my board has been stable ever since and faster than anything else I have. Any suggestions for an OC newbie who can't afford to simply keep trying different sticks until something works?Myrandex - Tuesday, July 12, 2005 - link
An interesting comparison would be an Athlon64 w/ DDR500 and a close to equivilently clocked Athlon64 w/ the HTT running at 250 to make the memory DDR500.semo - Tuesday, July 12, 2005 - link
#32, man if i could get those here in the uk (at that price).so Zebo, what do you mean that you can't run a64 in sync with memory. for example: if i get those:
http://www.ebuyer.com/customer/products/index.html...
and a dfi lanparty ultra d with a 3000 venice, would i be able to get a decent overclock?
Zebo - Monday, July 11, 2005 - link
It's really a shame anand did'nt do 3-3-3 like 99% of PC4200 sold runs @ 250...PrinceGaz - Monday, July 11, 2005 - link
Quite right Zebo, what this shows is that even the dual-core A64 processors when running multiple threads get little benefit from faster memory even at the same tight timings. Also as you say there is no such thing as a synchronous memory speed that might provide a performance advantage, as all memory speeds are in reality a divider from the CPU core speed. Running good PC3200 memory at or as much as it'll go above DDR400 2-2-2 is probably the best option. And if you want 2GB, get 2x 1GB modules so you can still run them at 1T command-rate as that's a better bet than four single-sided 512MB modules.One thing to bear in mind is if you enable Cool 'n' Quiet, that the memory may actually run faster at the lower CPU multipliers when it is set to other than DDR400 in the BIOS (DDR400 ensures the memory divider is always equal to the CPU multiplier).
Zebo - Monday, July 11, 2005 - link
http://forums.anandtech.com/messageview.aspx?catid...512 sticks
www.newegg.com/Product/Product.asp?Item=N82E16820146532
1024 sticks
www.newegg.com/Product/Product.asp?Item=N82E16820146545
#31 -- sure at a price, in the case of UTT is a pretty serious one if you ask me.. high volts.. high noise from fans.. high cost.. which may not be best usage of funds.. maybe that $150 saved is better spent on 7800GTX instead of 6800Ultra..or something like that.
AdamK47 3DS - Monday, July 11, 2005 - link
I'm a long time overclocker so a couple percent faster means something. If the option is there and it runs stable then go for it. All these little percentages add up in a heavily tweaked system. Overclocking the video, CPU, memory, and bus can all add up.Zebo - Monday, July 11, 2005 - link
PS I recommend Crucial sticks (not the 8T).. they have micron G abord..same as Ballistix for half price (but not speed binned so no guarantees).. Search around though.. many many including myself hit 2-2-2 with both 1024 x 2 and 512 x 2 configs.Zebo - Monday, July 11, 2005 - link
1. Keep in mind anand kept 2-2-2, low latency up to 250Mhz and *still* saw little difference.. fact is only one type, well two if you include anceint but still the best BH-5, can run this bandwidth and latency, Windbond UTT. This ram seems failure prone, watchout! The volts required, eg 3.5+, to attain those LL and bandwidth are extremely dangerous without active cooling on the ram. This ram is also "untested" (UTT) from windbond adding in an extra layer of uncertainty...Add in 2-3x the cost of regualar value muskin/corsiar/OCZ/Crucial which *can* all hit 2-2-2@200 with ~2.8V make this choice pretty lame considering the marginal payoffs. If you like cyber olympics and compete on the margins 0-5% by all means go for it...but 99% of y'all could find much better application of your funds.2. All other high bandwidth ram run crappy timings @ 250Mhz which will get stomped by LL @ 200. I've shown in forums you basically need 100mhz extra (not even 50) to hang with 200Mhz 2-2-2 when running ram 3-4-3, aka loose timings. Not worth buyin the high bandwidth stuff either when value muskin/corsiar/OCZ/Crucial which *can* all hit 2-2-2@200 with ~2.8V.
3. Overclcokers who want to run 1:1... there is no such thing as 1:1 in A64 archtecture.. all memory run async. So no problemo, no performance hit using 166/150/133 memsetting with value muskin/corsiar/OCZ/Crucial and shooting for 2-2-2@200 while clocking CPU to high heaven with proper HTT/FSB adjustments.
4. We see why AMD is'nt dieing for DDR2 this last year+.. and *when* they finally jump on board it will be at 667 instead of 533Mhz.. They can't afford the performance hit 533 will give, I'd be very surprised if 667 DDR2 is faster for AMD unless you run Crucials which is capable of 3-2-2.
In sum -- don't believe the hype. Get good cheap safe X2/A64 and DDR -- you'll be fast for a very long time.
robster3323 - Monday, July 11, 2005 - link
Is it possible to measure this stuff on a multi socket board? One of AMD's big selling points of HyperTransport is that other CPU's in other sockets can get to other memory faster. I wonder what impact these faster memories would have on socket to socket memory access?In other words socket one is direct connected to mem slot 1, the data in mem slot one is needed by a CPU in socket two, transiting the Hypertransport. How much benefit do the faster memories present?