The IGP Chronicles Part 1: Intel's G45 & Motherboard Roundup
by Anand Lal Shimpi & Gary Key on September 24, 2008 12:00 PM EST- Posted in
- Motherboards
Competitive Integrated Graphics?
Since the G45 GMCH is built on a 65nm process, it can be larger than G35's GMCH - and thus Intel increased the number of unified shader processors from 8 to 10.
| 1024 x 768 | Intel G45 | Intel G35 |
| Enemy Territory: Quake Wars | 6.8 fps | 5.6 fps |
| Company of Heroes | 24.5 fps | 16.5 fps |
| Race Driver GRID | 3.7 fps | 2.8 fps |
| Age of Conan | 7.9 fps | 6.1 fps |
| Crysis | 9.3 fps | 7.9 fps |
| Spore | 10.8 fps | 9.7 fps |
| Half Life 2 Episode Two | 40.7 fps | 27.9 fps |
| Oblivion (800 x 600) | 22.7 fps | 14.7 fps |
These shader processors are nearly directly comparable to NVIDIA's, meaning that in terms of raw processing power the G45 GMCH has 1/24th the execution resources of NVIDIA's GeForce GTX 280. It gets even worse if you compare it to a more mainstream solution - take the recently announced GeForce 9500 GT for example, it only has 32 SPs - putting the G45 at around 1/3 the clock-for-clock power of a 9500 GT.
Then there's the clock speed issue. While the GeForce 9500 GT runs its array of SPs at 1.4GHz, Intel runs its shader processors at 800MHz. Both Intel and NVIDIA's architectures have a peak throughput of one shader instruction per clock, so while the 9500 GT has 3x the execution resources of the G45 GMCH, it also has 75% clock speed advantage giving it a 4.6x raw throughput advantage over the G45 GMCH.
But how about comparing Intel's graphics core to NVIDIA's IGP equivalent? The GeForce 8200 is NVIDIA's latest integrated graphics core, it has 8 SPs and runs them at a clock speed of 1.2GHz - giving NVIDIA a 20% advantage in raw throughput on paper.
There are many unknowns here however. NVIDIA has special execution units for transcendentals and it's unclear whether or not Intel has the same. There are also times when Intel must relegate vertex processing to the CPU, which can cause strange performance characteristcis. But the point is that Intel's latest graphics core, at least on paper, is competitive to what NVIDIA is offering.
Neither the Intel or NVIDIA solution can hold a candle to AMD's 780G, which has a peak throughput of 40 shader operations per clock compared to 10 for Intel and 8 for NVIDIA. The reason AMD can do so much more is that each of its 8 processing clusters is 5-wide, just like on its desktop GPUs. If there's enough parallel data to work on, each one of these clusters can output five shader instructions per clock. The real world utilization is somewhere between one and five depending on how efficient AMD's real time compiler is and the code being run, but this generally translates into AMD dominating the IGP performance charts even with a lower clock speed than both Intel and NVIDIA parts.
Does this all really matter?
This next point is one that I've quietly argued for the past few years. ATI and NVIDIA have always acted holier than thou because of their IGP performance superiority over Intel, but I argue that they are no better than the boys in blue.
Despite both ATI and NVIDIA being much faster than Intel, the overall gameplay experience delivered by their integrated graphics solutions is still piss poor. Even on older games. Try running Oblivion, a 2.5-year old title, on even AMD's 780G and you'll find that you have to run it at the lowest visual quality settings, at the lowest resolutions (800 x 600, max) to get playable frame rates. At those settings, the game looks absolutely horrible.
In those games that aren't visually demanding, performance doesn't actually matter and all three vendors end up doing just fine. Fundamentally both ATI and NVIDIA want to sell more discrete cards, so they aren't going to enable overly high performance integrated solutions. The IGP performance advantages in games amount to little more than a marketing advantage, since anyone who actually cares about gaming is going to be frustrated even by their higher performing integrated solution.
The area where ATI/NVIDIA deliver where Intel historically hasn't is in the sheer ability to actually run games. In the past, driver issues and just basic compatibility with most games was simply broken on Intel hardware. Intel tried to address much of that with G45.
There is one aspect of IGP performance that really matters these days however: video decode acceleration.
Facebook
Twitter
RSS
53 Comments
View All Comments
Olyros - Sunday, February 8, 2009 - link
I noticed there's a mention in the article about searching for the perfect mini-itx case. The Nexus Psile case that I'm using for my Intel "DG45FC"-based computer is perfect for it I think. Especially if you are after stylish and quiet computers.Here are a couple of links for you to check it out if you want:
www.psile.com
http://www.3dgameman.com/forums/showthread.php?t=4...">http://www.3dgameman.com/forums/showthread.php?t=4...
lubama - Tuesday, October 14, 2008 - link
Quote from "Intel's G45 Motherboard round up"However, at various times after the system has gone to sleep it will wake back up without intervention for a few seconds and then shuts down. Sometimes this a few minutes after entering sleep mode, other times it occurred an hour or so later. The board requires a full power cycle to come back to life and does not always resume to Vista, instead we receive the error that Vista has been incurred an error after entering the OS."
Have you found a solution to this problem, seems like you are the only person, other than me, who is catching this issue. I have posted in numerous DG45ID forums and this exact issue is non-existent and haven't received any answers.
IntelUser2000 - Sunday, October 12, 2008 - link
Anand, G965/G35/GM965 has 8 EUs(Execution Units), but each of the EUs contain 2 cores, meaning it has 16 cores. Each cores can also process 2 threads, meaning it has a maximum of 32 thread capability. From that, its not comparable to Nvidia, nor ATI so Intel have their own performance metric.For G45, I assume its 10 EUs, 20 cores. Intel papers mention 50 threads.
puddnhead - Wednesday, October 8, 2008 - link
I second (thrid? fourth? fifth? 9th?) the call for the part 2 (what it's pretty obvious everyone is more interested in anyway, not this article).I wonder if you coudl at least give us an ETA of not the article itself? You know, if it's this week, this month, or ??? Thta doesn't seem too much to ask, I'm surprised you don't give that from the start.
computerfarmer - Friday, October 3, 2008 - link
Looking forward to part_2.I hope they are sooner than the promised reviews from these articles.
AMD SB750 arrives on the Foxconn A79A-S...
Date: July 21st, 2008
Author: Gary Key
http://www.anandtech.com/weblog/showpost.aspx?i=47...">http://www.anandtech.com/weblog/showpost.aspx?i=47...
AMD's SB750: Enabling Higher Phenom Overclocks?
Date: July 23rd, 2008
Topic: CPU & Chipset
Manufacturer: AMD
Author: Gary Key
http://www.anandtech.com/cpuchipsets/showdoc.aspx?...">http://www.anandtech.com/cpuchipsets/showdoc.aspx?...
AMD 790GX - The Introduction
Date: August 6th, 2008
Topic: CPU & Chipset
Manufacturer: AMD
Author: Gary Key
http://www.anandtech.com/cpuchipsets/showdoc.aspx?...">http://www.anandtech.com/cpuchipsets/showdoc.aspx?...
computerfarmer - Friday, October 3, 2008 - link
When is PART_2 coming out?duploxxx - Friday, October 3, 2008 - link
Lots of talk of bringing a great roundup of chipsets...already for a few weeks now.Where does anand start? at the least interesting and the most garbage chipset for several years now.
lets hope your global review is as good as people expect it to be.
you started off already a bit better then you did in recent gpu reviews, you actually took a cpu that was rather common to be used, although a e7200 or Q8200 would be a much better fit for this kind of boards.
whosthere - Tuesday, September 30, 2008 - link
I didn't see any specifics on the Game Settings. Could you please post them...Thanks
dutchroll - Monday, September 29, 2008 - link
Yeah I was thinking much the same thing about the "fanboi" comments.Can they spell "hypocrite"? It really betrays your allegiance when you rant at Intel then go all wobbly and weak at the knees while mentioning AMD. AT are damned if they do, damned if they don't as far as reviewing either brand's offerings. They've already stated how good the 780G was. They've stated what bugs are in the G45. So what the heck is the problem?
piesquared - Wednesday, October 1, 2008 - link
The problem is, they promised a SB750 review 2 months ago. The problem is they promised a DFI LP JR. review 1 month ago. The problem is they go well out of there way to avoid any comparison of Intel's ITD, and AMD's IGP. So what if they obscure and bury a line inside an Intel article that gives credit to AMD hardware. They fail to give credit where credit is due, and it is glaringly obvious. And it's even more obvious when this article pops up suddently when Intel has a new driver. They were waiting on Intel's promises of a new driver that would improve performance, and show it's hardware in a better light. Doesn't matter if you crap in a plastic or paper bag, it's still a bag of shit though. I'll make a wager that none of the upcomming "promised" reviews will have any side by side comparisons of Intel's IDT, to any other IGP. Unless of course AT stalls long enough to allow Intle more time to produce yet another driver.....