GeForce4 MX benchmarks

135

Comments

  • Reply 41 of 86
    zozo Posts: 3,117member
    for beeing the latest and greatest that Apple has to offer, compared to a PC with similar specs, the FPS that Apple reports are very poor.



    Specs like that would have me, and will have me, buy a PC for gaming.



    Q3 came out TWO YEARS ago... its an OLD game technology wise.



    And for your info, CounterStike is based on the HalfLife engine, which in turn is a HIGHLY modified Quake1+QuakeWorld+GL Quake, and itsy bitsy part of Quake 2 based engine. Nowadays HL engine is remotely based on original Quake engine.



    Still one of the best engines out there too imo. The realism it offers to games (Day of Defeat is amazing) is unparalleled to any other engine I have played with.
  • Reply 42 of 86
    macaddictmacaddict Posts: 1,055member
    [quote]Well, you could just as well ask if it would make sense to put a gamer chip into a pro-range computer.<hr></blockquote>



    The GeForce cards are more than just gaming cards. They have great 3D performance, but are good in other areas too. Is there some other non-gaming card you'd rather see in the PowerMac?



    [quote]No, in fact I was just curious as to what was necessary to get them to that point.<hr></blockquote>



    Stock cooling, the P4s seem to get to around 2.6GHz-2.8GHz using the 2.2GHz rated CPU. With water cooling etc, 3GHz is not unattainable. One guy got his P4 to 3.4GHz or so, using liquid nitrogen though. I can dig up the link if you want.



    [quote]Depends on what you do with it. For gaming, x86 PCs are obviously a better choice. Other areas might yield different results, especially considering stuff like 2MB L3 cache or GBit ethernet which are not found in most PCs.<hr></blockquote>



    Agreed. You can add a PCI card with gigabit ethernet though...it's not exactly a Mac-only feature.



    [quote]Sure. I was hoping someone might ask. I was hoping it would be Scott H, but oh well.



    <a href="http://duxbury.la/dp867b/dp867b_1.html"; target="_blank">http://duxbury.la/dp867b/dp867b_1.html</a>;



    The dual 867MHz G4 breaks 330fps.<hr></blockquote>



    He was using a home-made settings config that oddly enough more than doubled his performance, from 148 fps without this "patch" to 337 fps with. I find it slightly hard to believe that with his settings you can double your performance without any loss in quality.



    He also didn't seem to state what color depth he was running it at. Correct me if I'm wrong though on any of these points (as I'm sure you will be happy to do).



    I would download his settings myself, but they seem to be for the Cocoa version, and alas, I don't have OS X.
  • Reply 43 of 86
    idogcowidogcow Posts: 111member
    [quote]Originally posted by ZO:

    <strong>

    Still one of the best engines out there too imo. The realism it offers to games (Day of Defeat is amazing) is unparalleled to any other engine I have played with.</strong><hr></blockquote>



    Riiiiiight....



    Beats RtCW anyday
  • Reply 44 of 86
    Games are the best test of a system's performance. Mathmatica, Photoshop, Final cut pro all test the processor but games test everything. It needs the processor, video card, memory, hard drive. Everything is stressed. Games like Quake 3 give us a good understanding of how balanced a system is.



    Good for Apple if they have the newest Geforce 4 and high tech processors. But they aren't going to give you their max potential when restriced by slow hard drives, sdram and pathetic bus speeds.



    For $3,000 most people expect a system that can both work and play. Why should I have to dish out another $2,000 for a pc because Apple is incapable of adopting "old" technology?
  • Reply 45 of 86
    idogcowidogcow Posts: 111member
    Right, but for $1699 you can get a pretty decent PowerMac system with a Gforce4MX (oh, but wait, I forgot, the Gforce2MX is actually better )



    For Apple high end systems, they really need some help. (at least the prices went down a bit.)
  • Reply 46 of 86
    cobracobra Posts: 253member
    I think games will have problems on the Mac regardless of the the speed of the computer. Why?



    1. Game is not originally coded on the Mac.



    2. Drivers are probably not as mature on the Mac.



    3. Open GL performance is still not as up to snuff on the Mac as compared to the PC.





    I think if you took the same game and went totally independent on both , there would not be a huge difference.
  • Reply 47 of 86
    macaddictmacaddict Posts: 1,055member
    1. Quake 3, which we are mostly discussing, was coded for the Mac. In fact, the Q3 test came out for the Mac before the PC!



    2. Drivers for the...videocards? Well, that's something that we have to deal with. It can't be used as an excuse...kind of like saying "You're a better basketball player than me, but only because you practice".



    3. Doesn't Apple do it's own OpenGL drivers?
  • Reply 48 of 86
    [quote]Originally posted by MacAddict:

    <strong>



    He was using a home-made settings config that oddly enough more than doubled his performance, from 148 fps without this "patch" to 337 fps with. I find it slightly hard to believe that with his settings you can double your performance without any loss in quality.



    He also didn't seem to state what color depth he was running it at. Correct me if I'm wrong though on any of these points</strong><hr></blockquote>





    Believe it. I am able to improve fps on my G4 by a factor of two in timedemos simply by eliminating features such as dynamic lighting and adjusting the sound code, without loss in visual quality of the textures or polygons.



    These optimizations compounded with a 60-80% boost from the second processor pushes the machine over 300fps.



    BTW, He was running at 32bit depth.



    If you'd like, I can post the configuration file I use with my single G4 500 (OC'd from 450) w/Radeon. I achieve 80.4 fps under OS 10.1.2 at 800x600 resolution.



    [ 01-28-2002: Message edited by: Nostradamus ]</p>
  • Reply 49 of 86
    [quote]Originally posted by Nostradamus:

    <strong>





    Believe it. I am able to improve fps on my G4 by a factor of two in timedemos simply by eliminating features such as dynamic lighting and adjusting the sound code, without loss in visual quality of the textures or polygons.



    These optimizations compounded with a 60-80% boost from the second processor pushes the machine over 300fps.



    BTW, He was running at 32bit depth.



    If you'd like, I can post the configuration file I use with my single G4 500 (OC'd from 450) w/Radeon. I achieve 80.4 fps under OS 10.1.2 at 800x600 resolution.



    [ 01-28-2002: Message edited by: Nostradamus ]</strong><hr></blockquote>



    Ooooh! Gimme, gimme, please. I'm tired of 20fps on my radeon 32 ddr.
  • Reply 50 of 86
    This is for OS X Quake 3 only and works reliably only for single processor machines due to its idiosyncratic optimizations. Save the following text as a q3config.cfg file and place it in /Users/your_username/Library/Application_Support/Quake3/baseq3 directory .



    [code]

    // generated by quake, do not modify

    unbindall

    bind TAB "+scores"

    bind ENTER "+button2"

    bind ESCAPE "togglemenu"

    bind SPACE "+moveright"

    bind + "sizeup"

    bind - "sizedown"

    bind 1 "weapon 1"

    bind 2 "weapon 2"

    bind 3 "weapon 3"

    bind 4 "weapon 4"

    bind 5 "weapon 5"

    bind 6 "weapon 6"

    bind 7 "weapon 7"

    bind 8 "weapon 8"

    bind 9 "weapon 9"

    bind = "sizeup"

    bind [ "weapprev"

    bind ] "weapnext"

    bind _ "sizedown"

    bind ` "toggleconsole"

    bind c "+movedown"

    bind d "+back"

    bind g "+button3"

    bind r "messagemode2"

    bind t "messagemode"

    bind z "+zoom"

    bind ~ "toggleconsole"

    bind COMMAND "+moveup"

    bind PAUSE "pause"

    bind LEFTARROW "+left"

    bind RIGHTARROW "+right"

    bind ALT "+moveleft"

    bind CTRL "+attack"

    bind SHIFT "+speed"

    bind MOUSE1 "+forward"

    seta g_allowVote "1"

    seta g_filterBan "1"

    seta g_banIPs ""

    seta g_logSync "0"

    seta g_log "games.log"

    seta g_warmup "20"

    seta g_teamForceBalance "0"

    seta g_teamAutoJoin "0"

    seta g_friendlyFire "0"

    seta capturelimit "8"

    seta g_maxGameClients "0"

    seta sv_maxclients "8"

    seta timelimit "15"

    seta fraglimit "0"

    seta dmflags "0"

    seta g_logfile "games.log"

    seta color "red"

    seta cg_lagometer "0"

    seta cg_fov "90"

    seta cg_drawfps "1"

    seta cg_drawGun "1"

    seta cg_drawammowarning "0"

    seta cg_deferPlayers "0"

    seta cg_drawAttacker "0"

    seta cg_drawRewards "0"

    seta cg_drawSnapshot "0"

    seta cg_drawStatus "1"

    seta cg_drawtimer "1"

    seta cg_drawTeamOverlay "1"

    seta cg_bobup "0"

    seta cg_bobpitch "0"

    seta cg_bobroll "0"

    seta cg_runpitch "0"

    seta cg_runroll "0"

    seta s_loadas8bit "1"

    seta cg_simpleItems "0"

    seta cg_gibs "1"

    seta cg_forceModel "0"

    seta cg_draw3dIcons "0"

    seta server16 ""

    seta server15 ""

    seta server14 ""

    seta server13 ""

    seta server12 ""

    seta server11 ""

    seta server10 ""

    seta server9 ""

    seta server8 ""

    seta server7 ""

    seta server6 ""

    seta server5 ""

    seta server4 ""

    seta server3 ""

    seta server2 ""

    seta server1 ""

    seta cg_marks "0"

    seta cg_drawCrosshairNames "1"

    seta cg_drawCrosshair "5"

    seta cg_brassTime "0"

    seta ui_browserShowEmpty "0"

    seta ui_browserShowFull "1"

    seta ui_browserSortKey "0"

    seta ui_browserGameType "0"

    seta ui_browserMaster "2"

    seta g_spSkill "4"

    seta g_spVideos ""

    seta g_spAwards ""

    seta g_spScores5 ""

    seta g_spScores4 ""

    seta g_spScores3 ""

    seta g_spScores2 ""

    seta g_spScores1 ""

    seta ui_ctf_friendly "0"

    seta ui_ctf_timelimit "30"

    seta ui_ctf_capturelimit "8"

    seta ui_team_friendly "1"

    seta ui_team_timelimit "20"

    seta ui_team_fraglimit "0"

    seta ui_tourney_timelimit "15"

    seta ui_tourney_fraglimit "0"

    seta ui_ffa_timelimit "0"

    seta ui_ffa_fraglimit "20"

    seta com_soundMegs "8"

    seta s_buffersize "16384"

    seta s_chunksize "2048"

    seta s_mixPreStep "0.05"

    seta s_mixahead ".1"

    seta s_khz "11"

    seta s_doppler "0"

    seta s_separation "0.5"

    seta s_musicvolume "0"

    seta s_volume "0.8"

    seta r_appleTransformHint "1"

    seta r_maxDisplayRefresh "0"

    seta r_minDisplayRefresh "0"

    seta r_lastValidRenderer "ATI Radeon OpenGL Engine"

    seta cg_shadows "0"

    seta r_primitives "0"

    seta r_railSegmentLength "32"

    seta r_railCoreWidth "6"

    seta r_railWidth "16"

    seta r_facePlaneCull "1"

    seta r_gamma "1.306250"

    seta r_swapInterval "0"

    seta r_textureMode "GL_LINEAR_MIPMAP_LINEAR"

    seta r_finish "0"

    seta r_dlightBacks "1"

    seta r_dynamiclight "0"

    seta r_drawSun "0"

    seta r_fastsky "1"

    seta r_ignoreGLErrors "1"

    seta r_flares "0"

    seta r_lodbias "0"

    seta r_lodCurveError "250"

    seta r_ignoreFastPath "0"

    seta r_smp "0"

    seta r_subdivisions "4"

    seta r_vertexLight "0"

    seta r_simpleMipMaps "1"

    seta r_customaspect "1"

    seta r_customheight "1024"

    seta r_customwidth "1600"

    seta r_mode "4"

    seta r_ignorehwgamma "0"

    seta r_overBrightBits "1"

    seta r_depthbits "24"

    seta r_stencilbits "0"

    seta r_stereo "1"

    seta r_colorbits "32"

    seta r_texturebits "16"

    seta r_detailtextures "0"

    seta r_roundImagesDown "2"

    seta r_picmip "1"

    seta r_ext_texture_env_add "1"

    seta r_ext_compiled_vertex_array "1"

    seta r_ext_multitexture "1"

    seta r_ext_gamma_control "1"

    seta r_ext_compressed_textures "0"

    seta r_allowExtensions "1"

    seta r_glDriver "opengl32"

    seta cg_viewsize "100"

    seta cg_predictItems "0"

    seta cl_anonymous "0"

    seta sex "male"

    seta handicap "100"

    seta color2 "5"

    seta color1 "5"

    seta g_blueTeam ""

    seta g_redTeam ""

    seta team_headmodel "orbb"

    seta team_model "crash"

    seta headmodel "crash"

    seta model "crash"

    seta snaps "40"

    seta rate "40000"

    seta name "Fast config"

    seta cl_maxPing "800"

    seta m_filter "0"

    seta m_side "0.25"

    seta m_forward "0.25"

    seta m_yaw "0.026"

    seta m_pitch "0.022000"

    seta cg_autoswitch "0"

    seta r_inGameVideo "0"

    seta cl_allowDownload "0"

    seta cl_showmouserate "0"

    seta cl_freelook "1"

    seta cl_mouseAccel "0.3"

    seta sensitivity "30"

    seta cl_run "1"

    seta cl_packetdup "0"

    seta cl_maxpackets "40"

    seta cl_pitchspeed "9999"

    seta cl_yawspeed "140"

    seta sv_master5 ""

    seta sv_master4 ""

    seta sv_master3 ""

    seta sv_master2 ""

    seta sv_floodProtect "1"

    seta sv_maxPing "0"

    seta sv_minPing "0"

    seta sv_maxRate "0"

    seta sv_hostname "Renice"

    seta vm_ui "2"

    seta vm_game "2"

    seta vm_cgame "2"

    seta vid_screen "-1"

    seta in_mouseHighEndCutoff "0"

    seta in_mouseLowEndSlope "1"

    seta r_fullscreen "1"

    seta com_introplayed "1"

    seta logfile "0"

    seta com_blood "1"

    seta com_maxfps "400"

    seta com_hunkMegs "96"

    seta cg_zoomfov "22.5"

    seta cg_stereoSeparation "0.4"

    seta cg_draw2D "1"

    seta cg_drawIcons "0"

    seta cg_crosshairSize "24"

    seta cg_crosshairHealth "1"

    seta cg_crosshairX "0"

    seta cg_crosshairY "0"

    seta cg_railTrailTime "400"

    seta cg_teamChatTime "3000"

    seta cg_teamChatHeight "0"

    seta cg_drawFriend "1"

    seta cg_teamChatsOnly "0"

    seta cg_noVoiceChats "0"

    seta cg_noVoiceText "0"

    seta cg_cameraOrbitDelay "50"

    seta cg_scorePlums "0"

    seta cg_smoothClients "0"

    seta cg_noTaunt "0"

    seta cg_noProjectileTrail "1"

    seta ui_smallFont "0.25"

    seta ui_bigFont "0.4"

    seta cg_oldRail "1"

    seta cg_oldRocket "1"

    seta cg_oldPlasma "1"

    seta cg_trueLightning "0.5"

    seta cm_playerCurveClip "1"

    seta com_zoneMegs "16"



    </pre><hr></blockquote>



    [ 01-28-2002: Message edited by: Nostradamus ]</p>
  • Reply 51 of 86
    russsrusss Posts: 115member
    At the beginning of this topic Scott H. asked:

    [quote]Is it just me or do these benchmarks suck. I'm guessing it's not the GF4s fault.

    <hr></blockquote>

    Well, as far as I'm concerned there is nothing wrong with those benchmarks. See the revised chart below. I've included results from three of my computers for comparison.







    Now I would have to admit that the Radeon cards in the Athlon PC's must be just about maxed out, considering that there is only one fps difference between them. Unfortunately I don't have two to three hundred dollars for a new video card to find out.



    [ 01-29-2002: Message edited by: RussS ]</p>
  • Reply 52 of 86
    [quote]Originally posted by MacAddict:

    <strong>

    The GeForce cards are more than just gaming cards. They have great 3D performance, but are good in other areas too. Is there some other non-gaming card you'd rather see in the PowerMac?

    </strong><hr></blockquote>



    No, for most non-gaming work, the GF4MX is pretty good (especially given the dual monitor option). I was just pointing out that, for non-gaming stuff, using the "real" GF4 instead wouldn't help at all.





    [quote]<strong>

    Agreed. You can add a PCI card with gigabit ethernet though...it's not exactly a Mac-only feature.

    </strong><hr></blockquote>



    Well, it sort of is nonetheless - hardly any standard PC you buy today will come with 64-bit-PCI, and GBit on 32-bit-PCI doesn't make a lot of sense.



    Bye,

    RazzFazz
  • Reply 53 of 86
    [quote]Originally posted by RussS:

    <strong>At the beginning of this topic Scott H. asked:



    Well, as far as I'm concerned there is nothing wrong with those benchmarks. </strong><hr></blockquote>



    Look at some of the other benchmarks we've linked to.
  • Reply 54 of 86
    serranoserrano Posts: 1,806member
    <a href="http://www.barefeats.com/pentium4.html"; target="_blank">benchmarks here</a>



    also i found out theres a command to enable smp in quake 3 over at arstechnica



    [quote]Originally posted by strepidus:

    I'm just hoping that when Apple did the new dual 1GHz tests with Q3A and reported an fps of 115, they had forgotten to enable symmetric multiprocessing (r_smp 1). I mean, even a dual 800 with a GF3 can get over 120 fps with everything maxed out at 1024x768. So unless the GF4 MX really sucks, the dual 1GHz box should be doing much better than "just" 115 fps.<hr></blockquote>



    <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />



    that about sums it up, how does the 7500 compare, why isnt the gf3 still offered, when is the 8500 coming... because apparently this gf4mx just can't cut it... someone get kidred to test his dualie out in q3 with smp on...



    [edit]<a href="http://www.apple.com/powermac/graphics.html"; target="_blank">heres</a> where i got the 115 from, in case anyone hasn't seen it already[/edit]



    [ 01-29-2002: Message edited by: janitor ]
  • Reply 55 of 86
    Well graphics cards are not my area. I don't keep up with what's out and what's fastest. When I saw those benchmarks my first thought was "That sucks. Why show those horrible numbers?"



    [ 01-29-2002: Message edited by: Scott H. ]</p>
  • Reply 56 of 86
    macaddictmacaddict Posts: 1,055member
    [quote]Believe it. I am able to improve fps on my G4 by a factor of two in timedemos simply by eliminating features such as dynamic lighting and adjusting the sound code, without loss in visual quality of the textures or polygons.<hr></blockquote>



    You can't compare tests done at highest quality vs. something else without skewing it. I don't believe any of these tests run on PCs compromised viewing quality. I don't seem to have an option for dynamic lighting...do you mean vertex vs. lightmap?



    [quote]These optimizations compounded with a 60-80% boost from the second processor pushes the machine over 300fps.



    BTW, He was running at 32bit depth.



    If you'd like, I can post the configuration file I use with my single G4 500 (OC'd from 450) w/Radeon. I achieve 80.4 fps under OS 10.1.2 at 800x600 resolution.<hr></blockquote>



    ...wish I had OS X.



    How do you know he was at 32bit? Just curious.
  • Reply 57 of 86
    sebseb Posts: 676member
    I guess it's all about perspective.



    From my perspective those are great benchmarks.



    If you're comparing them to a machine that's made specifically for gaming they may not be as good as that machine.



    It would take a hell of a lot more than only getting 115+FPS in a game to make me consider getting a PC.



    I think Apple would have to go away for me to get a PC...even then I'd probably just make my Mac last as long as possible.



    Just my perspective.



    Here's another perspective (just for fun):



    I mentioned to my gf that some people were complaining that new Macs didn't get as high of a FPS than some PCs for games, but that they were more than triple what I get now. Her response was, "if someone is getting a computer just to play games, they probably have bigger problems. They probably need to get outside a little more". heh. Doesn't apply to everyone, but I thought it was a funny perspective, nonetheless.
  • Reply 58 of 86
    Lemme bring it back the most important point.



    Apple's $3000 top of the line system shows less performance than a PeeCee at much lower price. You pay top dollar for the Mac and get mediocre performance. That hurts!
  • Reply 59 of 86
    macaddictmacaddict Posts: 1,055member
    [quote]"if someone is getting a computer just to play games, they probably have bigger problems. They probably need to get outside a little more"<hr></blockquote>



    They just don't get it. <img src="graemlins/oyvey.gif" border="0" alt="[No]" /> <img src="graemlins/lol.gif" border="0" alt="[Laughing]" />
  • Reply 60 of 86
    applenutapplenut Posts: 5,768member
    [quote]Originally posted by RazzFazz:

    <strong>



    Well, it sort of is nonetheless - hardly any standard PC you buy today will come with 64-bit-PCI, and GBit on 32-bit-PCI doesn't make a lot of sense.



    Bye,

    RazzFazz</strong><hr></blockquote>



    what "standard" PC are you paying 3000 dollars for?
Sign In or Register to comment.