Intel Yonah Performance Preview - Part I: The Exclusive First Look at Yonah
by Anand Lal Shimpi on November 30, 2005 2:50 AM EST- Posted in
- CPUs
Yonah vs. Dothan
We didn’t have much time to put together this piece, but at the same time we wanted to present the most complete picture of Yonah as possible, so we went back to our last Pentium M on the desktop article and configured our Yonah system identically so we’d have as close to an apples-to-apples comparison as possible. Of course it is impossible to use the same motherboard, due to the socket differences we’ve already mentioned, but the rest of the systems are configured identically. We apologize in advance for the brevity of the benchmark suite, in due time we will present an even more thorough look at Yonah, but for now we are working with what we’ve got. Also keep in mind that the platform and processor are both pre-release samples, so performance could change, most likely for the better.
With that said, we've got a question and that is: how does Yonah stack up to Dothan?
Unfortunately, our Yonah only runs at 2.0GHz, and our reference Dothan numbers are from a 2.13GHz CPU - so we don’t get the clock for clock comparison we were hoping for, making it even more difficult for Yonah to impress. Thankfully our first benchmark is clock speed independent as we look at how cache latencies have changed from Dothan to Yonah using ScienceMark 2.0:
| L1 Cache Latency | L2 Cache Latency | |
| Dothan | 3 cycles | 10 cycles |
| Yonah | 3 cycles | 14 cycles |
And changed they have indeed. If you’ll remember from our earlier desktop Pentium M investigations, Dothan’s very quick 10 cycle L2 cache allowed it to be competitive with AMD’s Athlon 64, despite lacking an on-die memory controller. With the move to Yonah however, the L2 cache latency has gone up a whopping 40%. While we’re still dealing with a lower access latency than the Pentium 4, this increase will hurt Yonah.
We’re guessing that the increase in access latency is due to the new dynamically resizable L2 cache that’s used in Yonah. In order to save power as well as maximize the use of the shared L2 cache between cores, Yonah can dynamically adjust the size of its L2 cache, flushing data to main memory when faced with low demand. The associated logic is most likely at least partially to blame for the increase in L2 cache latency.
So Yonah has a slower L2 cache working against it, but two cores and a handful of architectural enhancements working in its favor - let’s see how they stack up in the real world.
First up, we’ve got our business application tests:
| Business Winstone 2004 | Communication (SYSMark 2004) | Document Creation (SYSMark 2004) | Data Analysis (SYSMark 2004) | |
| Dothan (2.13GHz) | 24.3 | 129 | 202 | 118 |
| Yonah (2.0GHz) | 21.6 | 146 | 215 | 138 |
Dothan has a sizeable lead in Business Winstone 2004, which we’ve always attributed to its low latency L2 cache. Since the benchmark gets no benefits from dual core, and doesn’t take advantage of any of the SSE improvements to Yonah, the advantage is clearly in Dothan’s court.
The SYSMark tests paint a different picture, with Yonah outpacing the faster clocked Dothan by 6 - 17%. What’s interesting to note is that in these tests, the performance advantage isn’t exclusively attributable to the advantage of having two cores - Yonah’s architectural advancements are at work here as well.
The digital content creation tests are where Yonah’s improvements should shine:
| Multimedia Content Creation Winstone 2004 | 3D Content Creation (SYSMark 2004) | 2D Content Creation (SYSMark 2004) | Web Publication (SYSMark 2004) | |
| Dothan (2.13GHz) | 29.8 | 188 | 255 | 169 |
| Yonah (2.0GHz) | 34.7 | 264 | 323 | 236 |
And shine they do; thanks to a combination of the move to dual core as well as the architectural improvements over Dothan, Yonah shows anywhere between a 16 - 40% increase in performance.
| DivX | Doom 3 | |
| Dothan | 39.7 fps | 95.5 fps |
| Yonah | 57.5 fps | 93.8 fps |
The DivX test shows what we’ve pretty much seen across the board from dual core scaling in video encoding, so there’s no surprise there. Our only gaming benchmark, Doom 3, shows a hazier picture with Dothan on top, and Yonah close behind. We will investigate gaming performance of Yonah much closer later on.
What we can walk away from these benchmarks with is an idea of the level of improvement to expect from Yonah, but now comes the real test - how does it stack up against other desktop processors, especially the Athlon 64 X2.
Facebook
Twitter
RSS
135 Comments
View All Comments
bhtooefr - Wednesday, November 30, 2005 - link
Also, isn't there a way to disable the second core in BIOS?I think it's definitely possible to make your dual-core chip into a single-core for testing.
Slaimus - Wednesday, November 30, 2005 - link
No, since it is a CPU with double execution units, rather than 2 CPUs linked together.fitten - Wednesday, November 30, 2005 - link
I've not seen that stated anywhere. What I've seen is that there are two cores, two independent L1 caches, and a shared L2 cache.Anemone - Wednesday, November 30, 2005 - link
The reason it was compared against the X2 is simply because that is where the next gen Turion "may" land in performance. Power consumption in this test is nice, but not really relevant to the new Turion, nor is it relevant to the mobile arena where AMD with integrated controller does well against P-M plus external controller consumption (well though Intel is better so far).I had a feeling things were going to turn out this way. That cache on the P-M is really what makes it such a nicely performing chip. Weakening that was really the telling test in all this. It's nice that you put that test first. The dual core Turion is going to be very close in release timing to Yonah, and quite a few samples of the chip are already out there, but I'd bet AMD is keeping its cards close, as it has less of a need to prove a point.
However, from a mobile perspective AMD's use of Via and ATI chipsets really will be the telling factor, and will impact the "you get it all" bundle of a notebook. Where, after over a year, is Nvidia and a pci-e mobile chipset for AMD? Turion is almost out the door and they haven't even spoken up yet. That's probably because they won't be there when Turion releases. If Nvidia had no chipset on the desktop, the competition would be starkly different right now, and a good many folks know this to be true. So I'm expecting the dual core Turion to do incredibly well. 64 bit isn't an end all be all, but it will be a wonderful perk when Vista comes, and Vista will be here sooner than previously expected. Power consumption "may" favor Yonah, however Intel is doing what they always do, when a chip performs so-so, they push up the clockspeed. When they find out what Turion has on the table, they'll put out the EE version of Yonah to try to match it. THAT I bet, won't be so power friendly.
Intel is improving on Yonah with Conroe. But remember that the low latency L2 is what gave the Dothan its ability to stand in the A64 arena. That got lowered and we see the results. Now imagine what will happen when they "gently" increase the pipeline of Conroe vs P-M/Yonah. It'll drop a bit again. Against that you'll have architecture improvements and higher clockspeed (Conroe is expected to clock higher). That might end up making Conroe only "slightly" better than Yonah. That is why the marketing of th two can overlap, probably.
So IF (and it's not promising) Nvidia gets the NF4 into a mobile format and gets builders to use it on the new Turion dual cores, we might see some interesting things from the AMD side. If not, they stay as now, a niche market only. ATI chipsets aren't going to change that, much as they have not done much to the desktop side of things.
IF Yonah clocks at 2.6ghz you may see some competitive desktop systems to A64 X2's. It will rule the mobile roost unless the above happens for AMD. Conroe had better make some dramatic improvements to even Yonah's ability, though I'm doubtful, or putting both Yonah and Conroe on the desktop will simply hand even more of the market to AMD. On the mobile front, again unless the above happens for AMD/Nvidia on a mobile chipset, you will watch Yonah/Conroe rule the mobile roost.
:)
Griswold - Wednesday, November 30, 2005 - link
Intel just now managed to close the gap between them and AMD as far as dual core is concerned. I dont care if one is a mobile part and the other is not, mobile computing is of no interest to me and even less dual core mobiles. But it gives a good idea of what to expect.The numbers are there. Yonah isnt groundbreaking besides the fact that it is a dual core - and the desktop versions probably wont be either. Wow, thats late...
nserra - Wednesday, November 30, 2005 - link
TG Daily: What is your reason for not including support for 64-bit technology in Yonah? Why is it going to be supported only in Meron?Perlmutter: Like everything in technology or life, it is all a matter of timing. Integrating unneeded features into a processor means a waste of power consumption. Our assumption was, and still is, that 64-bit extensions are not the most important thing required from a processor, not even in the beginning of 2006. We believe this will change with Windows Vista, when applications start migrating to 64-bit, and we want to be ready then, and not a minute earlier.
Furen - Wednesday, November 30, 2005 - link
64-bit extensions = more transistors, which means higher power consumption.Donegrim - Wednesday, November 30, 2005 - link
If this is cheaper than the X2 3800, and overclocks really well on the desktop (say, 2.7-3ghz or so), then I could see myself being tempted by it. As long as the 'secret' motherboard isn't as ridiculously overpriced as current dothan desktops.Griswold - Wednesday, November 30, 2005 - link
WHat makes you think that thing will go to 2.7-3ghz? *laff*Donegrim - Wednesday, November 30, 2005 - link
Well the x2 3800 that consumes much more power (therefore is hotter) consistently goes to at least 2.5ghz, so I don't think it's unrealistic for a lower power chip to go further. Plus the FSB is nice and low, which makes for a good start with the overclocking.