"Intel Nehalem benchmarked" (Hexus)
- Thread starter AmberClad
- Start date
2.93 ghz vantage score on nehalem cpu is 17,000
2.93 ghz vantage score on yorkfield cpu is 10,200
Game framerates are a mixed bag, kinda disappointing. I don't see this thing being faster than a 4.5ghz penryn in games. (unless it oc's that high as well)
2.93 ghz vantage score on yorkfield cpu is 10,200
Game framerates are a mixed bag, kinda disappointing. I don't see this thing being faster than a 4.5ghz penryn in games. (unless it oc's that high as well)
Vantage shows a 70% increase with Nehalem clock for clock - that doesn't downplay Nehalem, IMO. It shows it shining bright. 
I was referring to the actual gaming benchmarks.
I was referring to the actual gaming benchmarks.
Looks to be a solid performer for sure. Is anyone here pretty familiar with HyperThreading? I remember it was overall a benefit on the P4's as I understand it because it took a weakness of that chips design and turned it into a strength. Because of the long pipeline, instead of leaving it stalled at times it could work on two streams of data to keep everything busy, as I understood it. Does anyone know how many stages are in the C2D or Nehalem processor? I thought they were supposed to be fairly short, so how does HyperThreading work, or work well? I'm no expert, so I may be off on what I said, but that's more or less how I understood the P4's to use HyperThreading, so any education on the subject would be appreciated.
z1ggy
Lifer
- May 17, 2008
- 10,010
- 66
- 91
Originally posted by: Rubycon
Yawn. Vantage (anything that Futuremark publishes actually) should be ignored. Encoding bench show promise just not enough to warrant jumping ship (awful term btw) to this platform. Give me a 22nm Haswell instead.
Ok in four years?
Jared, unless a game is CPU bottlenecked, what would adding a more powerful CPU do to framerates? That is not a good benchmark......
Originally posted by: SlowSpyder
Looks to be a solid performer for sure. Is anyone here pretty familiar with HyperThreading? I remember it was overall a benefit on the P4's as I understand it because it took a weakness of that chips design and turned it into a strength. Because of the long pipeline, instead of leaving it stalled at times it could work on two streams of data to keep everything busy, as I understood it. Does anyone know how many stages are in the C2D or Nehalem processor? I thought they were supposed to be fairly short, so how does HyperThreading work, or work well? I'm no expert, so I may be off on what I said, but that's more or less how I understood the P4's to use HyperThreading, so any education on the subject would be appreciated.
Hyperthreading on Nehalem is actually functioning how SMT was originally invented to function, making use of idle resources on the processor. Processors generally have execution units going unused (as opposed to clock cycles) so the unused function units get used with hyperthreading.
Looks like replacing the 6MB L2$ per 2 cores with the 0.5MB L2$ per core + IMC results in higher memory latencies.
I recall the C2D latency is well hidden by those intentionally mammoth L2$ sizes coupled with the purportedly uber aggressive prefetchers...but apparantely things do take a step back here with the significant reduction in L2$ sizes.
Not that it will make a drastic performance impact on Nehalem, just means that it would stand to reason that Westmere is going to have more L2$ per core to gain back some of that latency masking advantage that C2D's prefetchers and L2$ sizes brought to the table.
I recall the C2D latency is well hidden by those intentionally mammoth L2$ sizes coupled with the purportedly uber aggressive prefetchers...but apparantely things do take a step back here with the significant reduction in L2$ sizes.
Not that it will make a drastic performance impact on Nehalem, just means that it would stand to reason that Westmere is going to have more L2$ per core to gain back some of that latency masking advantage that C2D's prefetchers and L2$ sizes brought to the table.
Interesting observation, though it seems the reviewer is confident a fully tweaked production level Nehalem will at least match the latency of the QX9770.
Yeah I saw that comment too. I am very leery of any reviewer who openly admits they are on some sort of misson to showcase a new CPU in the best possible light they can contrive.
The bottom line is that the latency is higher for the L1$ and the L3$ has a 40cycle latency that must be dealt with before going to the IMC. (same problematic issue that K10 deals with)
Nemesis 1
Lifer
- Dec 30, 2006
- 11,366
- 2
- 0
Well I have read alot of your post. One aspect of a more poweful CPU would be physics If intel does it threw the CPU. Also watch the demoes raytracing. Larrabbee I am fairly certain will be working real hard along side Nehalem in doing physics and RT.
ALSO every review I have seen so far. Shows Nehalem to be slower than penryn in games. What the reason for this really is unknown its also strange . Either Nehalem is slower in gamesOR intel is holding back . But considering the smaller L2 cache it just simply may be slower. One must remember on the top end their is that QP link. I wonder if this is kin to AMD's side port.
Originally posted by: SlowSpyder
Looks to be a solid performer for sure. Is anyone here pretty familiar with HyperThreading? I remember it was overall a benefit on the P4's as I understand it because it took a weakness of that chips design and turned it into a strength. Because of the long pipeline, instead of leaving it stalled at times it could work on two streams of data to keep everything busy, as I understood it. Does anyone know how many stages are in the C2D or Nehalem processor? I thought they were supposed to be fairly short, so how does HyperThreading work, or work well? I'm no expert, so I may be off on what I said, but that's more or less how I understood the P4's to use HyperThreading, so any education on the subject would be appreciated.
Pipeline depth doesn't really have much to do with SMT suitability. The point of HT was to keep idle execution units busy by giving schedulers more instructions to pick from, improving efficiency.
Core 2/i7 with is much wider exeuction engine stands to benefit greatly from SMT, but only in cases where the thread number exceeds the core number!
I must say that I'm very dissapointed with Nehalem's single-threaded and gaming performance. Since I'm primarily a gamer, this may mean that I'll wait until Nehalem is released, then get a cheaper q9650, overclock it, and wait to see what Westmere has to offer.
I wouldnt be drawing too many conclusions from that article.....I wouldnt take either the Vantage or the gaming seriously.
Yah thought so.
"As IDF has started, the first benchmarks of Nehalem will probably pop up. It is without a doubt an impressive architecture that gets a much better platform to run on, but this CPU is not about giving you better frames per second in your favorite game than the Penryn family. Let me make that more clear: even when the GPU is not the bottleneck, it is likely that most games will not be significantly faster than on Penryn. We, the people behind it.anandtech.com will probably have the most fun with it, more than your favorite review crew at Anandtech.com
. And no, I have not seen any tests before I type this. Nehalem is about improving HPC, Database and virtualization performance, much less about gaming performance. Maybe this will change once games get some heavy physics threads, but not right away.Why? Most Games are about fast caches and super integer performance. After all, most of the Floating point action is already happening on the GPU. All Core 2 CPUs were a huge step forward in integer performance (not in the least because of memory disambiguation) compared to the CPUs of that time (P4 and K8). Nehalem is only a small step forward in integer performance. And the gains due to slightly increased integer performance are mostly negated by the new cache system. In a previous post I told you that most games really like the huge L2 of the Core family. With Nehalem they are getting a 32 KB L1 with a 4 cycle latency, next a very small (compared to the older Intel CPUs) 256 KB L2 cache with 12 cycle latency and after that a pretty slow 40 cycle 8 MB L3. When running on Penryn, they used to get a 3 cycle L1 and a 14 cycle 6144 KB L2. That is a 24 times larger L2 than Nehalem!
The percentage of L2 caches misses of most games running on a Penryn CPU is extremely low. Now that is going to change. The integrated memory controller of Nehalem can't help much, as the fact remains that the L3 is slow and the L2 is small.
But that doesn't mean Intel made a bad choice. Intel made a superbly good choice by improving the performance where Core (Merom/Penryn) was mediocre to good. Penryn was already a magnificent gaming CPU, but it could not beat the AMD competitor in HPC benchmarks. And AMD gave good resistance in the database performance benchmarks. That is all going to change.
Most Database code can not use the wide architecture of Penryn very well. The number of instructions per cycle get lower than 0.5 and waiting for the memory is the most probably cause. SMT or Hyperthreading can do wonders here: while one thread waits for a memory stall, the other thread continues working and vice versa.
Secondly, quad (and eight) socket performance is going to improve a lot as four Nehalems only have to keep four L3 in sync, while a similar Tigerton system has to keep 8 L2 caches in sync. That is why the cache system is perfect for server performance, but a little less interesting for gaming performance.
The massive bandwidth that the integrated tri-channel memory controller delivers will do wonders for HPC code. And the new TLB architecture with EPT will make Nehalem shine compared to it's older Core brothers.
No, Nehalem was made to please the IT and HPC people. Bring it to it.anandtech.com, it is not that interesting for you gamers ;-)
"
Hl2 Ep2 2560x1600 8xAA/4xAA 16xAF:
Tech Report (4xAA):
Testbed:
Core 2 Extreme QX9650 3.0GHz
4870X2 CFX: 128.2
GTX 280 Tri SLI: 108.2
http://techreport.com/articles.x/15293/6
Driverheaven (8xAA):
Testbed:
2x Intel QX9775 Engineering Samples (@ 4.4 Ghz ? 11x400, 1.475 volts)
4870X2 CFX: 165
GTX 280 Tri SLI: 171
http://www.driverheaven.net/re...reviewid=609&pageid=10
- Sep 28, 2005
- 21,067
- 3,575
- 126
okey i promised i would keep my nose out of this thread.
But b4 it gets derailed with more speculation.
Guys, you havent seen the 3.2 EE yet.
You havent seen the potential in overclocking yet.
You guys are basing judgements on a late end beta product, probably one of the first sample batches to hit.
Your calling judgements b4 the retail has hit.
Stop trying to pull a rabbit out of the hat. The rabbit isnt fully in there yet.
This is what im trying to tell you guys.
But b4 it gets derailed with more speculation.
Guys, you havent seen the 3.2 EE yet.
You havent seen the potential in overclocking yet.
You guys are basing judgements on a late end beta product, probably one of the first sample batches to hit.
Your calling judgements b4 the retail has hit.
Stop trying to pull a rabbit out of the hat. The rabbit isnt fully in there yet.
This is what im trying to tell you guys.
apoppin
Lifer
No not really getting you analogy.
Anand made it pretty damn clear that Neha is for everyone *except* gamers - me. He is telling us the color of the rabbit.
so .. What ARE you trying to tell us?
--that Neha does impress for gaming?
--that it is so impressive - also o/c'd - over Penryn's current O/C'd 4+Ghz that we should wait for it?
my 4870x2 is arriving this week and i am going for e0 q550 as soon as i get back from Nvision and x48 .. that is the plan, so far.
Well why would purely gamers need 4 cores with HT right now anyway.
Hopefully by the time i7 is supposed to be "mainstream" (2H 2009) there are some games that can use quads. (SupCom and FS:X, yes I know).
I dont know why a "gamer" would possibly want more than a high-clocked E8XXX for the next 12-18 months.
Hopefully by the time i7 is supposed to be "mainstream" (2H 2009) there are some games that can use quads. (SupCom and FS:X, yes I know).
I dont know why a "gamer" would possibly want more than a high-clocked E8XXX for the next 12-18 months.
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: 22K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 411
Facebook
Twitter