GeForce GTX 780M, 770M, And 765M: Scaling Vs. Radeon HD 8970M

Benchmarking Configurations

Using the stats reported by GPU-Z, this is how our hardware and software is being benchmarked. We had to make a couple of small changes to Eurocom's notebook as it was delivered, though we didn't have to mess with its cooling system. We’re relying on both Origin PC's and Eurocom’s best assembly efforts to show off the highest possible performance from each piece of hardware.

Test System Configuration
CPUIntel Core i7-4930MX: 3 to 3.9 GHz, 8 MB Shared L3 Cache, FCPGA946
Motherboard, ChassisClevo P177SM: Intel HM87 Express, 4 x DIMM, 3 x SATA/2 x mSATA/1 x eSATA 6Gb/s, HDMI, Dual DisplayPort, 17.3" FHD 1080p
Cooling SystemDual-blower air: 2 x CPU pipes, 2 x GPU pipes, 1 x GDDR5 pipe
RAMKingston 99U5469-035.A00LF (8 GB)
2 x 4 GB DDR3-1333 CAS 9-9-9-24, Dual-Channel Mode
Nvidia GraphicsNvidia GeForce GTX 780M: 771-797 MHz GPU, 4 GB GDDR5-5000
Nvidia GeForce GTX 770M: 706-797 MHz GPU, 3 GB GDDR5-4008
Nvidia GeForce GTX 765M: 797-863 MHz GPU, 2 GB GDDR5-4008
AMD GraphicsAMD Radeon HD 8970M: 900 MHz GPU, 4 GB GDDR5-5000
Hard DriveSamsung 840 Pro MZ-7PD256, 256 GB SSD
SoundIntegrated HD Audio
NetworkRealtek 802.11b/g/n + Bluetooth v4.0+LE Combo Half Mini-Card module
PowerChicony A12-230P1A: 100-240 VAC to 19.5 VDC, 11.8 A
System Software
OSMicrosoft Windows 7 Home Premium x64
Nvidia GraphicsNvidia GeForce Mobile 332.21 WHQL
AMD GraphicsAMD Catalyst Mobility 13.12

The top two GPUs in this line-up are capable of pushing playable performance at far higher resolutions than the 1920x1080 supported by each notebook’s panel. We’ve also seen a couple gamers tie their notebooks to larger displays. Fortunately, all four GPUs are able to fully implement the P177SM’s DisplayPort outputs.

StarTech's MDP2DVID DisplayPort-to-dual-link DVI adapter supplies the bandwidth to feed our aging Dell 30" screen its native 2560x1600 resolution. DisplayPort-enabled QHD monitors might be more modern, but QHD is still a step down in resolution from this behemoth.

3D Game Benchmarks
Arma 3
Version 1.08.113494, 30-Sec. Fraps "Infantry Showcase"
Test Set 1: Standard Preset, No AA, Standard AF
Test Set 2: Ultra Preset, 8x FSAA, Ultra AF
Battlefield 4Version 1.0.0.1, DirectX 11, 100-Sec. Fraps "Tashgar"
Test Set 1: High Quality Preset, No AA, 4X AF, SSAO
Test Set 2: Ultra Quality Preset,  4X MSAA, 16X AF, HBAO
Far Cry 3V. 1.05, DirectX 11, 50-sec. Fraps "Amanaki Outpost"
Test Set 1: High Quality, No AA, Standard ATC., SSAO
Test Set 2: Ultra Quality, 4x MSAA, Enhanced ATC, HDAO
F1 2012Steam version, in-game benchmark
Test Set 1: High Quality Preset, No AA
Test Set 2: Ultra Quality Preset, 8x AA
Metro: Last LightSteam version, Built-In Benchmark, "Frontline" Scene
Test Set 1: DX11, Med Quality, 4x AF, Low Blur, No SSAA, No Tesselation, No PhysX
Test Set 2: DX11, High Quality, 16x AF, Normal Blur, SSAA, Tesselation Normal, No PhysX
Tomb RaiderSteam version, Built-In Benchmark
Test Set 1: High Quality Preset (8x AF, FXAA), Motion Blur, Screen Effects
Test Set 2: Ultimate Quality, (16x AF,  FXAA), Tesselation, TressFX
Synthetic  Benchmarks
3DMark ProfessionalVersion 1.1, SystemInfo 4.17.0.0, Fire Strike Benchmark (Extreme Off/On)
Create a new thread in the US Reviews comments forum about this subject
This thread is closed for comments
43 comments
    Your comment
    Top Comments
  • Gaming laptop, there's an oxymoron if I ever heard one. The "twist" in the value analysis changes nothing. Hell I could play with numbers to prove that an Intel i7-4960X is a "better value" than an AMD FX-6300 just by dropping it into a $10,000 system. Playing the percentages instead of the actual dollars makes it look better than it actually is because even though the percentage is smaller, it's still a smaller percentage of a larger number and ends up being the same. You say that the 8970M is 33% less expensive while the 780M is 13% faster? Man, you really do know how to screw with numbers don't you? The 780M is indeed 13% faster than the 8970M but it is a FULL 50% MORE EXPENSIVE than the 8970. Why are you trying so hard to word it like nVidia is the better value? No tech journalist is stupid enough to do it by accident!
    13
  • Nice review.
    I can honestly that I was not expecting Pitcairn to perform so well against GK104 while also maintaining lower power consumption!
    10
  • Other Comments
  • Nice review.
    I can honestly that I was not expecting Pitcairn to perform so well against GK104 while also maintaining lower power consumption!
    10
  • Are Optimus or Switchable Graphics available in these systems?With those solutions around, I wish we could start seeing reasonable (>4-5 hr) battery life out of these laptops when not under load.
    0
  • In about a year or so when there is a catalog of Mantle games, laptops could make a comeback as a viable gaming platform under $1500.
    0
  • When will they make mainstream external boxes for these gaming cards? Give it a cheap and fast link to a notebook and price it competitively to the gaming laptops; awesome upgrades like the desktop crowd.
    0
  • I own a 780M.I would like to see 780M versus 8970M using Mantle in BF4.
    1
  • I really like my 8970M. It's awesome having that much power in a notebook. I just hope it doesn't die like my 6990M did.
    1
  • A great place to check GPUs in laptops and laptops in general is notebookcheck.netIts especially important because heat can be a problem in laptops very, very fast.Sure those did not seem to have that problem but ambient temps, as well as dust acumulation are two factors rearly taken into consideration when talking about overheating.
    0
  • I have seen people buy the kind of laptops that these cards exist in. The main issue with them is that they are often so compromised in one way or another they spend more time getting sent back for repairs and fixes than they do on the owners desk.Like TVR cars, they did more miles strapped on the back of low loaders going to and from Blackpool than actually on its own wheels.
    0
  • the wife has a MSI i7QM 3630 i believe and a 670mx. It has been over a year and still runs strong. Alpha testing EQNLandmark with her laptop and it runs like a champ on a mix of ultra and high settings. we have not had one issue with it and runs pretty dang cool and quiet for a gaming laptop. not sure what machine the people you have seen Daglesj, but we are very happy with ours.
    0
  • Wish you included the GT750m SLi available on the Lenovo Y510P. I own said laptop and $ for $ couldn't ask for more, but have no real goal post against the other chips in the Nvidia range.Great to finally see some mobile GRFX chop reviews and look forward to more in the future.
    0
  • Well, speaking about reliability, I own a alienware m17r4 with a 675m (580m slightly overclocked) never had to be repaired and still running strong. I often use it for gpu rendering tasks which are the equivalent of running furmark for several hours. Cpu never went below the turbo frequency gpu never throttled or got above 70c. The system is one and a half years old now and still kicking :-)
    0
  • Oh and btw optimus is by default enabled on the above tested systems. You get an astonishingly long battery life with these systems (given the specs)... If you have enough muscles to carry them around :-)
    0
  • What do you mean actual temperature minus ambient? What is the ambient temperature? If it is a 20 celcius then gtx 780m is 92.6 celcius??? That is very hot if it is true!
    0
  • I like the new twist in the value analysis, where you determine the overall system price at which the different components add the most value. It would be good to see this incorporated into the standard value analysis for desktop components as well.
    0
  • There seems so little development in the laptop sector, they consistently artificially rig the market to make consumers spend around the £800 mark for anything capable of running pc games (look at the dell revamp of the inspiron 17 which is around the £1000 mark). I know there is the Lenovo but why don't more manufacturers produce machines in a 17.3 form factor with those specs? It seems you have a choice of badly specced machines that are very cheap or really extortionately priced gaming laptops with little in between.
    0
  • Gaming laptop, there's an oxymoron if I ever heard one. The "twist" in the value analysis changes nothing. Hell I could play with numbers to prove that an Intel i7-4960X is a "better value" than an AMD FX-6300 just by dropping it into a $10,000 system. Playing the percentages instead of the actual dollars makes it look better than it actually is because even though the percentage is smaller, it's still a smaller percentage of a larger number and ends up being the same. You say that the 8970M is 33% less expensive while the 780M is 13% faster? Man, you really do know how to screw with numbers don't you? The 780M is indeed 13% faster than the 8970M but it is a FULL 50% MORE EXPENSIVE than the 8970. Why are you trying so hard to word it like nVidia is the better value? No tech journalist is stupid enough to do it by accident!
    13
  • The cost of the gpu alone may be 50% more but when you put the whole system into account you may end up paying it 33% more, not 50%. Beside that, this time the price/performance ratio seems to favor amd imho.
    0
  • I wish the performance of the desktop counterparts would have been added to the charts. or even the desktop PRICE counterparts, so the 780M compared to the 780TI etc.
    3
  • Blazer1985, what you didn't understand is the difference of point of view between the 2 comparisons. It says "8970M is 33% less expensive // 780M is 13% faster". It should be either :"780M is 50% more expensive // 780M is 13% faster"OR"8970M is 33% less expensive // 8790M is x% slower"
    2
  • Don't play with numbers! :) 500 to 750 is 50% more expensive, BUT 750 to 500 is 33% less expensive, damn the mathematics!
    2