Comments Locked

59 Comments

Back to Article

  • gbrux - Wednesday, May 18, 2005 - link

    I'd like to see a performance comparison when each configuration, 32-bit and 64-bit, includes the maximum amount of memory for the motherboard.

    It is reported that Windows 64 with AMD 64 handles large amounts of memory better than a 32-bit system, thereby producing better performance.

    Why did AT use just 1GB memory?

    So, test that.

  • magnusfl - Saturday, September 5, 2009 - link

    what i not see so far is any testing with more then 4 gigs of memory as I sure there a big jump in some games due to the limitation of 2 gig for the game which is not the case in the 64 bit OS
    as when the level is reached the memory overflow goes to the Hard drive which is much slower then memory
  • Carfax - Saturday, May 14, 2005 - link

    Anand may have underestimated how GPU bound 1024x768 at VERY HIGH quality is. The guy who claims the 46% increase used Medium quality settings, which may have exposed the processors influence a lot more.
  • Carfax - Saturday, May 14, 2005 - link

    Doug, I was just about to post that. We need more testers to verify that performance boost though, because it was done by an amateur.

    If it's true though (and I hope so) it bodes well for games which are optimized for AMD64.
  • dougSF30 - Friday, May 13, 2005 - link

    And here is a real comparison that actually makes sense:

    64b FarCry runs up to 46% faster than 32b.

    Gosh, who would've thought that not making the GPU the bottleneck would expose the code speed improvement?


    http://www.aceshardware.com/forums/read_post.jsp?i...


  • maynardc - Thursday, May 12, 2005 - link

    I have the XP Pro x64 disc and have a few questions:

    1. If I install XP Pro x64 over XP Home Edition, will I still be able to run all of currently installed 32-bit apps OR will I have to reinstall everything?

    2. Also, if I uninstall it, will it automatically go back to XP Home Edition SP2?
  • Concillian - Thursday, May 12, 2005 - link

    Who is the tard? The person who wrote the review, or the person who didn't read it well enough to realize that the performance comparisons weren't including the exclusive content that brought the increased visuals? Personally my vote is for the latter.

    "in order to isolate the performance differences from the extra content, we only looked at performance changes with the first 64-bit patch installed - not the Exclusive Content Update."

    Reading Comprehension is your friend.
  • islandtechengineers - Thursday, May 12, 2005 - link

    I thought the same when i saw it. the improved graphics are better than the 5 to 6% performance increase! visually there's a noticeable difference; I doubt anyone would notice a 5 or 6% on performance.
  • TWilliams - Thursday, May 12, 2005 - link

    Who ever wrote this article is a flipping tard!!! LOL... Look at it like this; The game runs 5% to 6% faster and it has greatly improved graphics! One would think we are comparing the same game and only getting 5 or 6 percent increase from the review, but that is not the case! We are getting 5 to 6 percent increase with greatly improved graphics, there is nothing to laugh at when you look at it like that... Any how, crappy review.. hehe
  • smn198 - Thursday, May 12, 2005 - link

    #25 I can't see how your reference to [H]ard|OCP helps your argument. See the final page. http://www.hardocp.com/article.html?art=NzY3LDY=

    "Our performance results show that we did not have an increase or a decrease in performance with 64-bit FarCry with the new content. This is actually quite reassuring considering that the ECU patch is actually adding in a lot of new content, including a new rendering feature called Offset Mapping.

    What CryTek and AMD have managed to do is to give us more content and world quality in the game without losing performance! They are taking the power of the AMD Athlon64 CPU and leveraging that to offset any performance differences and adding more detail in the game.

    What we found the most demanding on performance is the new Offset Mapping 3D technique. In Catacombs, which employs this feature all over the scene since you are basically surrounded by stones the entire time, we found performance here to drop the most compared to 32-bit FarCry without Offset Mapping. There were instances where the frame rate was lower by 10-15FPS with Offset Mapping compared to without. However, in most of those scenes, the frame rate was already very high at approximately 70FPS, so losing 10FPS off that wouldn’t cause any change in gameplay experience. Indeed, it seems where Offset Mapping is used, performance is slightly slower in those situations than without from our experiences. There are also scenes and maps though where performance can be slightly faster, a few FPS here and there with 64-bit FarCry, and that’s with the extra content.

    It seems to be both a give or take relationship. Depending on the scene or map, performance could be slightly slower, or slightly faster with 64-bit FarCry. Nevertheless, it is never too drastic to affect real world gameplay."

    In light of your link, I’ll amend my first comment from “#3,5,8 I am impressed. If I understood correctly this is not comparing apples to apples. There are performance gains whilst having higher detail. That is great!” to “#3,5,8 I am impressed. If I understood correctly this is not comparing apples to apples. There is negligible performance deviation whilst having higher detail. That is great!” Hope you are happy now.

  • Jynx980 - Thursday, May 12, 2005 - link

    More insects and birds? I better hop on the 64 bit wagon!

    Im surprised to see so many uneeded processes running. qttask, ati2evxx, ipodservice, ituneshelper; these things eat up the ram little by little.
  • KF - Wednesday, May 11, 2005 - link


    There is no way to say for sure that the basic design of Far Cry is sufficiently ambitious to show off AMD64 well, because designers need to design to what is available in mass if they plan to sell games in large quantity. Intel only recently capitulated to doing EMT64, so designers could not design with 64 bit in mind.

    It may be that AMD's intent was to show that more abitious game designs are more practical for a 64 bit machine. If AMD wants to demo advantages of AMD64, there needs to be a 32 bit version of the game with the extra content to compare, or else they need to show why the added content is not possible with 32 bits.

    Why do more registers speed things up? Keeping things in registers is the fastest way to do things. x86 has 8 general purpose registers, 6 of which are ordinarily available. To do a very simple loop, you are liable to need 2 or 3 registers to hold adresses and at least 2 to hold operands. Since you need to hold onto intermediate results in the process, you are already nearly maxed out. That was fine in the original 386 days. But current CPUs can readily execute 3 ops simulanaeously, so if you attempt to use the CPU to maximum capablity, and do multiple things concurrently, you run out of registers. Therefore the capabilities of present CPUs are vastly underutilized. Adding 8 more general registers, like AMD64 does, helps a lot. Intel says that by actual measuremnent in real programs, currrent CPUs average 1 op per cycle, when they can do a possible 3. Not impressive. I hope that explains the potential of more registers.

    OTOH, programs like games obviously use massive amounts of data which is only briefly in registers, so slinging that in and out of memory is going to take up a lot of time. That will limit the average ops per cycle considerably. It is only while you are processing a given chunk of data that you could get the number of ops per cycle up. The more you process a given chunk of data, the more potential speed up possible. If the design was not ambitious enough, there is not going to be much speed up. Older CPUs would be good enough.

  • thelanx - Wednesday, May 11, 2005 - link

    Has anyone compared the enhanced visual quality screenshots with hdr screenshots? I know that enabling hdr makes the colors of the land and the water more vibrant, which seems to occur in the enhanced version as well, most noticably the better water and more vibrant trees. Perhaps crytek used what they learned doing the hdr and implemented similar effects without using hdr?
  • Sheeno - Wednesday, May 11, 2005 - link

    I don't get the problem, Anand showed that the difference between the 64bit patch and a normal 32bit farcry was negligible. In fact, the only difference is the graphical enhancements you get from adding the second 'graphics' patch. We can all assume that the quality of your graphics card will run the latter, so why the 64bit? It seems this was nothing more then a marketing stunt, and those of us with 32bit processors, AMD or not, are left without a neat update. Thanks AMD, I'm glad you decided to forsake your 32bit processors.
  • ir0nw0lf - Tuesday, May 10, 2005 - link

    As much as I love AT, I feel that the [H]ard|OCP article was much more informative and "meaty." Their article didn't seem to have a rushed feel to it vs. the AT article, which as mentioned above feels a bit rushed and possible incomplete.
  • msva124 - Tuesday, May 10, 2005 - link

    Thank you, that makes sense.
  • stephenbrooks - Tuesday, May 10, 2005 - link

    ^ Compiler doesn't get in knots swapping stuff in and out of main RAM or cache so often. Registers have a 0 cycle access delay :p
  • msva124 - Tuesday, May 10, 2005 - link

    >That's like asking why would more cache, or more RAM increase performance..

    Then it should be very easy for you to explain why more registers will increase performance.
  • Concillian - Tuesday, May 10, 2005 - link

    I will go along with many of the opinions I'm seeing here and say that I think there is a lot of assumption in the article, and not enough proof.

    - The reader is assumed to know that 1024x768 high quality is CPU bound

    - The article assumed that the performance of 64 bit patch + additional content would show worse performance, but has no demonstration of that in either a CPU bound or GPU bound scenario. (speaking of the comments: "In fact, a fast GPU is all you really need to take advantage of most of these features - not a 64-bit CPU" and "but the fact of the matter is that none of the visual improvements enabled by the Far Cry patches had anything to do with AMD64 or EM64T"

    Perhaps those comments assume more knowledge of how all this stuff works than the average reader has, and it's more obvious to someone like Anand with a more extensive knowledge of CPU architecture and how software interacts with specific registers, etc... Maybe the conclusion is more obvious to someone with that kind of knowledge, but I just see conclusions being drawn about one thing (64 bit patch + ehanced content) from a test of a completely different thing (64 bit patch only)

    Anand, you talk about apples to apples comparison. To take that analogy and run with it... Why do you do an apples to apples comparison and then make a conclusion about potatoes?

    I think it would have been a reasonable comparison to see the performance with and without the additional content to demonstrate which hardware was actually being taxed (video or CPU). Especially with the strong opinion expressed in the article about what hardware the etra content taxes. I don't question the conclusion necessarily, I just think that if you maake such a strong statement you should have real data to back it up.

    I only criticize because we have high expectations of AT articles. I hope these comments are seen as constructive, because that's the intent.
  • xtknight - Tuesday, May 10, 2005 - link

    I agree that some additional benchmarks with the 64-bit content pack would have been a good idea, even if it's not apples to apples. I'm not downloading this patch for the 5 FPS increase, I want the extra eye candy. Is the content pack supposed to include only stuff a 64-bit CPU could do efficiently? Then it would be nice to compare it to a 32-bit CPU. Otherwise, I understand your reasoning that this pack really has nothing to do with 64-bit in the first place.
  • Son of a N00b - Tuesday, May 10, 2005 - link

    good article anand, i loved your quick and to the point style on this one. Any qestion i had you just cleared up.

    keep it up, thats why I come here.

    If i have more questions then i read more reviews, that simple
  • Backslider - Tuesday, May 10, 2005 - link

    #9 I was refering to Far Cry not AMD.

    Their special content idea blows goats, it pisses me off royal. Their game sucks anyway, so no real loss here.

  • robg1701 - Tuesday, May 10, 2005 - link

    Anand,

    1. Like adding a faster cpu increasing performance in tasks that werent cpu bound ? Like the A64 getting higher numbers than a GPU bound P4 with the same card at the same settings ? Odd as it seems i dont think it can be so easily ruled out.

    2. Apples to oranges it may be, but thats exactly what the existance of 64bit far cry represents just now, and is exactly how i think anyone bothering to download the patches will play it, so in bothering to do an article on it i think it ghas to be examined. Now that other articles are online, so far your the only site ive seen not do this.

    3. I didnt imply they were SM3 specific, but its a difference between the 6800 and X850, and i think re-examining their relative performance is required given we now also have a new OS, new drivers, and new game code.

    But no matter, other sites have filled the void and benchmarked what I and many others think is the sensible comparison - waht people will actually be looking for when they download a gigabyte of updates.
  • jballs - Tuesday, May 10, 2005 - link

    [H]ard OCP reports that there are minimal diffrences in frame rates (some higher some lower) when the patch AND the additional content are applied (read conclusions). So, increased quality with no much impact on performance is always good thing. The thing that is still up in the air is what is the diffrence between 64 and 32bit processors with that additional content enabled because the game won't let you install on anything else. For all we know the performance would be the same.
  • nserra - Tuesday, May 10, 2005 - link

    #33 Finaly some light!!!!!

    Thanks!
  • PottyWithElmo - Tuesday, May 10, 2005 - link

    anandtech. sorry. (refer to 31)
  • Anand Lal Shimpi - Tuesday, May 10, 2005 - link

    nserra

    Sorry I should have been more specific - when I refer to the 64-bit version of Far Cry I am talking about both the binary patch and the content update (as both require a 64-bit OS and CPU to run).

    All of the benchmarks were done with only the first patch installed, which doesn't improve any of the features I mentioned in the image quality section.

    All of the screenshots were taken with both patches installed.

    Take care,
    Anand
  • Anand Lal Shimpi - Tuesday, May 10, 2005 - link

    dougSF30

    1) We don't test at 640x480 in our CPU shootouts, but I should note that the X850 XT happens to be CPU bound at 1024x768 and below.

    2) The impact of dual core on the comparison doesn't really change anything; performance would scale the same on single core CPUs.

    3) The reason for mentioning that the new content doesn't use more than 2GB of memory is because all of the enhancements to the game have nothing to do with a 64-bit CPU; you could just as easily implement any of these features on a 32-bit machine and have the same end result.

    With reference to what the patch we tested with enables, it looks like the first patch does not change any of the visual options which is why I felt comfortable making that comparison. All of the level changes are definitely made with the second patch.

    As far as the performance with the second patch applied; our default demo uses the mp_airstrip level which showed no performance difference with or without the extra content. At higher resolutions I saw some frame rate drops in certain levels but at more CPU bound resolutions there was basically no difference.

    robg1701

    1) That analysis would imply that we are GPU bound at 1024x768 in our test - which is not true.

    2) We didn't include performance numbers with the enhanced content applied simply because it wouldn't be an apples to apples comparison. The other issue with including those numbers is that the performance impact of the additional content really varies from one level to the next, as does the visual impact of the content.

    3) None of our information on the patches lead us to believe that any of the features were shader model 3.0 specific.

    Take care,
    Anand
  • PottyWithElmo - Tuesday, May 10, 2005 - link

    I believe FarCry 64bit runs more content (view distance, extra entities like bugs and birds, more complex textures) at a same or better framerate than 32bit, correct? How is this not a decent improvement over 32bit again? If, in 32bit, I turn on exra birds and bugs and increase the view distance (configure it in the game.cfg file of FCry), my framerate and performance drops significantly. This article should be torn up, new benches performed, and then rewritten. It makes no sense. It reads like its someone whose got a grudge with AMD, instead of looking at a FREE upgrade in farcry that adds extra content at a slightly better framerate! Just stick to the facts Annantech!
  • nserra - Tuesday, May 10, 2005 - link

    I already dont understand a cheat!

    Anand Lal Shimpi clear this up!

    The bench was made with no extra content but the pictures are taken with the extra content or not?

    "The differences in the Pier level are numerous. First, the improvement to draw distance is very noticeable:"

    "You can see additional detail in the rock textures:"


    OR I CAN READ MR MLITTL3 #14 OR ANAND LAL SHIMPI CANT WRITE!!!!

    "And finally, the water looks a lot better in the 64-bit version:"

    then must be repleaced with

    "And finally, the water looks a lot better with the AMD64 Exclusive Content Update:"
  • Cygni - Tuesday, May 10, 2005 - link

    The 64bit hardware DID make a difference in how the game was run. Look at the first chart. Not only was the game a little bit faster in 64/64... but the detail was increased a pretty hefty ammount. It wasnt some gimmick to make 64bit computing LOOK good by giving an exclusive and not delivering... 64bit computing actually HAD an impact, on both image quality and speed. Not bad.
  • jediknight - Tuesday, May 10, 2005 - link

    #26:
    (jaw drops)
    That's like asking why would more cache, or more RAM increase performance..
    (/jaw drop)
  • ncage - Tuesday, May 10, 2005 - link

    What people need to understand is these games were most likely NOT optimized for 64 bit. Yes the source code was compiled with a 64 bit compiler but probably not optimized. Those of us who are programmers know that if you want optimal speed then you want to use 32 bit numbers (integers, Floating point, ect) (exactly the size that can fit in a register).With 64 bit processors that changes of course. Even Tim Sweeny (probably spelled his name wrong) said he had to get around the limit of 32 bit registers. Now if we took UT2004 and just compiled it with a 64 bit compiler would it help that much? Umm, maybe a little. Now if Tim wouldn't had to get around the limitation of 32 bit registers and had 64 bit integers and floating point thats where we would have seen the difference. My point is that when people start designing their programs for 64 bit processors is when you will see the big difference. Just because they compile something with a 64 bit compiler doesn't mean its going to improve things that much.
  • msva124 - Tuesday, May 10, 2005 - link

    Why would having more registers increase performance?
  • mlittl3 - Tuesday, May 10, 2005 - link

    For all you whiners out there asking for more detail, go to

    http://www.hardocp.com/article.html?art=NzY3

    for a more indepth review. They list all the changes in the patch, tons of screenshots and show no difference between 32 and 64 bit.
  • dougSF30 - Tuesday, May 10, 2005 - link

    Yeah, there's also the silly complaining about not needing more than 2GB for the new content. 64b is about more than addressible memory, and any speed improvement has virtually nothing to do with larger memory, but rather, the increased number of registers, and operations on 64-bit data. So why go on about the memory footprint of the new content?
  • xTYBALTx - Tuesday, May 10, 2005 - link

    This article leaves me with more questions than answers.
  • Bonesdad - Tuesday, May 10, 2005 - link

    cool!! More birds and insects!!!...

    hrm...
  • robg1701 - Tuesday, May 10, 2005 - link

    Im not too sure about this review to be honest.

    First up, the comment regarding intel benfiting more than AMD, could it not just be that with the AMD scores being higher in the first place tehre is less room in the radeons performance capability for improvement ? Need more benchies to establish the reason i say..

    Next up, why no benches of the 64bit enhanced mode ? We have no way of knowing from this article if the enhanced 64bit content slows it down or not, which is quite ridiculous given thats one of if not the main reason for bothering to download the 64 bit patch.

    Finally, nvidia cards have historically been a bit slower in Far Cry in particular, but nvidias 64bit drivers have been doing the rounds for a lot longer sp perhaps they are more mature and can reap more benefit, wouldnt it be nice to benchmark an 6800 series card for those users who dont have an X850? Adding to this point, the 6800s having PS 3.0 gives them a different featureset to the X850 under Far Cry, does that make any difference here ?

    All in all, a bit of a rushed article I felt, if your going to bother doing smaller stuff like this at least do it a bit more whole heartedly ?
  • dougSF30 - Tuesday, May 10, 2005 - link

    That merely begs the question.

    What changes are made in the "binary" patch, and what changes are made in the "exclusive content update" patch.

    Draw distance is hardly "exclusive content", so it isn't clear from the name of each patch.

    Which patch causes it to be changed?

  • ViRGE - Tuesday, May 10, 2005 - link

    #16: Taken from the article "As we just implied, the 64-bit add-ons to Far Cry come in two separate packages. First, there's the actual 64-bit patch that installs and enables a native 64-bit binary to run under x64 edition. The second package is the AMD64 Exclusive Content Update that improves the actual content in the game."

    So AT only installed the 64-bit patch, which was the executable only, the test was not done with the content patch, and as a result the test was apples-to-apples.
  • dougSF30 - Tuesday, May 10, 2005 - link

    There may not be "extra content", but is the draw-distance increased?

    Other problems with this article, if one is looking to measure code speed improvement 32b vs. 64b:

    1) Should be run in low-detail, 640 x 480. Just like in CPU shoot-outs.

    2) Single-core CPUs should have been used.

    But again, the biggest *potential* issue is with any changes between 32b and 64b versions that do not fall into the "exclusive content" patch. Draw-distance is not content. What about other "all-level" changes? Are they affected by the first patch, or the second?
  • Icehawk - Tuesday, May 10, 2005 - link

    It would be interesting to see if the added content did make a change to the FPS #s though.
  • smn198 - Tuesday, May 10, 2005 - link

    #14 Glad it was clear for you. Where does it say in the article which changes are in the 64-bit patch and which are in the Exclusive Content Update?

    Reading this I was lead to believe that running the 64-bit patch changed textures, view distances, etc for everything in the 64-bit version.

    "AMD listed the changes to the 64-bit version of Far Cry as follows:

    All Levels

    * Improved terrain textures
    * Increased view distance
    * Offset bump mapping added for rock and stone objects
    * More insects and birds"
  • jediknight - Tuesday, May 10, 2005 - link

    I wonder if the video card drivers for XP64 are at par with their 32 bit counterparts.. I wonder if the performance advantage you get by going 64bit is negated by performance disadvantages in 64bit video drivers.
  • mlittl3 - Tuesday, May 10, 2005 - link

    #6, #10

    I wish you people would read the article. This is an apples to apples comparison. Please read the f*cking article.

    The article clearly states,

    "We used our standard Far Cry demo that we've used in all other reviews, and in order to isolate the performance differences from the extra content, we only looked at performance changes with the first 64-bit patch installed - not the Exclusive Content Update."

    The benchmarks are not including the extra content. Damn, and I thought PC enthusiasts could read. Read the article next time.
  • saechaka - Tuesday, May 10, 2005 - link

    did anyone see the difference between intel and amd? thats why i bought a amd64, and whatever benefits i get elsewhere is a perk
  • Lord Zado - Tuesday, May 10, 2005 - link

    #11, you don't need to justify your 64 bit processor because it likely allows all 32 bit games to run faster than on a 32 bit *cough* Intel *cough* processor. Now if you made the switch to 64 bit Windows already...
  • L3p3rM355i4h - Tuesday, May 10, 2005 - link

    #10, exactly! It made a good looking game that much better, IMHO, better than Doom3. And it still runs better. Granted, the engine has been matured, but its possible they could have tweaked the AI also to take advantage of the 64 bit power.

    But maybe i'm just trying to justify my 64 proc after all ;)
  • smn198 - Tuesday, May 10, 2005 - link

    #3,5,8
    I am impressed. If I understood correctly this is not comparing apples to apples. There are performance gains whilst having higher detail. That is great!
  • AnnihilatorX - Tuesday, May 10, 2005 - link

    #8

    You have to understand though, if AMD didn't put the money and effort in it these features won't be available at all.

    Plus they didn't charge you for extra for their CPU because they put some money into FarCry did they?
  • Backslider - Tuesday, May 10, 2005 - link

    "AMD's goals are quite admirable, but the fact of the matter is that none of the visual improvements enabled by the Far Cry patches had anything to do with AMD64 or EM64T - they are artificially limited to run on those platforms alone, but could work just as well on a 32-bit platform."

    This sickens me. Enable special things for people who bought special hard ware, EVEN THOUGH THE HARDWARE MAKES NO DIFFERENCE TO HOW IT RUNS!

    Thats just a pathetic ploy at the customer to convince them to adopt new hardware they dont need.

    Sad Really. See if I buy anything from these assholes.
  • Anemone - Tuesday, May 10, 2005 - link

    Nice to see ANY improvement, but it's tiny really. The painful detraction to this gain is that most 32 on 64 programs are running slower such that the gain in one is offset by losses in a dozen or more other apps.

    But we KNEW it would take years before the power of 64bit would become commonplace even having the hardware and an OS for the hardware. But I'm seeing nothing to make me compelled to switch yet, which kind of hurts the progress :(

    $.02
  • nserra - Tuesday, May 10, 2005 - link

    Well I’m impressed the image quality is superb, but I don’t know if the added image quality has more to do with the graphics card than to the processor.

    Amd and Ubi should have made the older "image quality" mode available just to compare.
  • rqle - Tuesday, May 10, 2005 - link

    nothing big, but the improvement is very much welcome.
  • AnandThenMan - Tuesday, May 10, 2005 - link

    #2, exactly right.

    Mildly impressed with the gains, better than nothing.
  • wien - Tuesday, May 10, 2005 - link

    Hmm.. That was mildly disappointing. I was sure the extra registers would have more of an impact than that.
  • weiran - Tuesday, May 10, 2005 - link

    Heh, can't get Safedisc working in 64-bit?
  • Pjotr - Tuesday, May 10, 2005 - link

    What a pity, still, 4-6 % improvements are nothing to sneeze at, some 200 GHz meaning you can opt for a speed grade lower CPU and save some money.

Log in

Don't have an account? Sign up now