AMD FirePro W600 vs NVIDIA GeForce GTX 670MX Comparison

AMD
RADEON

AMD FirePro W600

CORE STATE Cape Verde
VRAM 2 GB
CLOCK SPEED
TDP 75 W
BUS WIDTH 128 bit
ARCHITECTURE GCN 1.0
nm
PROCESS 28 nm
LAUNCH DATE 2012
VS
NVIDIA
GEFORCE

GeForce GTX 670MX

CORE STATE GK104
VRAM 3 GB
CLOCK SPEED 601 MHz
TDP 75 W
BUS WIDTH 192 bit
ARCHITECTURE Kepler
nm
PROCESS 28 nm
LAUNCH DATE 2012

PERFORMANCE BENCHMARKS

geekbench_opencl
6,223
6,125
geekbench_vulkan
N/A
5,316

Analysis: AMD FirePro W600 vs NVIDIA GeForce GTX 670MX

Head-to-Head Benchmarks

The database records a single head-to-head benchmark between these two GPUs: Geekbench OpenCL. In that test, the AMD FirePro W600 scores 6223, while the NVIDIA GeForce GTX 670MX scores 6125. The AMD part wins by 1.6%, a narrow margin that places both cards in essentially the same performance tier.

Context from the nearest rivals list reinforces this picture. The FirePro W600 sits within 0.2% of the AMD Radeon HD 6950 (6210) and trails the NVIDIA GeForce RTX 4070 Ti SUPER AD102 by just 0.7% (6270). The GeForce GTX 670MX, meanwhile, lands within 0.2% of the Intel Iris Pro Graphics P6300 (5712) and within 0.5% of the AMD Radeon HD 8790M (5691). Both cards occupy the 33rd to 36th percentile of all GPUs in the database, meaning they cluster around the same mid-range performance band.

The win count shows 1 for the FirePro W600 and 0 for the GTX 670MX, but that single benchmark does not tell the whole story. The GTX 670MX has an additional Geekbench Vulkan score of 5316, a test the FirePro W600 does not have recorded. When averaging all benchmark scores, the FirePro W600 achieves 6223, while the GTX 670MX averages 5721 across its two tests. That aggregate difference is about 8.8%, larger than the head-to-head OpenCL delta would suggest.

The recorded data indicates that the OpenCL result is close enough to be within measurement noise. A 1.6% delta between two cards from different architectures, with different memory configurations and different shading unit counts, suggests that raw compute throughput in this particular workload is nearly identical. The FirePro W600's slight edge could stem from its higher effective memory clock (4 Gbps vs. 2.8 Gbps) or its GCN 1.0 architecture's handling of OpenCL work, but the database does not provide per-workload breakdowns to confirm such hypotheses.

Where Each One Wins

The AMD FirePro W600 wins the only directly comparable benchmark, Geekbench OpenCL, by 1.6%. This result, combined with its higher average benchmark score (6223 vs. 5721), indicates that the FirePro W600 holds a measurable advantage in compute-oriented workloads that leverage OpenCL. Its 768.0 GFLOPS FP32 throughput, while lower than the GTX 670MX's 1,153.9 GFLOPS, does not translate into a loss in this test. The FirePro W600's memory bandwidth of 64.00 GB/s on a 128-bit bus, paired with GDDR5 running at 4 Gbps effective, appears sufficient to keep up with the GTX 670MX's 67.20 GB/s on a 192-bit bus.

The NVIDIA GeForce GTX 670MX's only recorded win is its Vulkan score of 5316, a benchmark that the FirePro W600 does not have in the database. This means the GTX 670MX demonstrates at least baseline Vulkan capability, while the FirePro W600's Vulkan support is listed as version 1.2.170 but untested. For users prioritizing Vulkan-based workloads, the GTX 670MX has the only concrete data point.

The GTX 670MX also holds advantages in raw specification counts. It has 960 shading units versus 512, 80 TMUs versus 32, and 24 ROPs versus 16. Its texture rate is exactly double: 48.08 GTexel/s versus 24.00 GTexel/s. These numbers suggest the GTX 670MX should excel in geometry-heavy or texture-bound tasks, even if the OpenCL benchmark does not capture that advantage. The pixel rates are nearly identical (12.02 vs. 12.00 GPixel/s), so no meaningful difference appears there.

The FirePro W600 counters with a higher memory clock (1000 MHz vs. 700 MHz) and higher effective data rate (4 Gbps vs. 2.8 Gbps). It also has a more substantial transistor density: 12.2M per mm² versus 12.0M per mm². These efficiency metrics hint that the FirePro W600 extracts more performance per transistor, despite having fewer total transistors (1,500 million vs. 3,540 million).

The Verdict

The data points to different buyers for each card. The AMD FirePro W600 is the choice for OpenCL compute workloads. Its head-to-head win, higher average score, and higher percentile (36th vs. 33rd) all favor it. The card's 6x mini-DisplayPort 1.2 outputs also make it suited for multi-display professional setups, though the database does not record a benchmark for that use case.

The NVIDIA GeForce GTX 670MX is the choice for Vulkan workloads, based on the only Vulkan score in the comparison. Its larger memory pool (3 GB vs. 2 GB) and wider memory bus (192-bit vs. 128-bit) provide more headroom for texture-heavy applications, even if its effective bandwidth is only slightly higher (67.20 vs. 64.00 GB/s). The GTX 670MX's 80 TMUs and 960 shading units suggest it could handle tasks that the OpenCL benchmark does not exercise, such as tessellation or advanced shader effects.

Neither card is a clear winner for gaming. Both are end-of-life products, and the database does not include gaming-specific benchmarks. The GTX 670MX's higher raw FP32 throughput (1,153.9 GFLOPS) and double texture rate give it a theoretical edge in game-like workloads, but no recorded data confirms this.

For a professional workstation environment prioritizing OpenCL, the FirePro W600 is the safer pick. For a portable device or a workload that requires Vulkan, the GTX 670MX is the only option with recorded evidence. The average benchmark gap of 8.8% in favor of the FirePro W600, driven by the GTX 670MX's lower Vulkan score, suggests that the AMD card delivers more consistent performance across the benchmarks it participates in.

FAQ

Q: Which GPU wins the Geekbench OpenCL benchmark?

A: The AMD FirePro W600 scores 6223, while the NVIDIA GeForce GTX 670MX scores 6125. The FirePro W600 wins by 1.6%.

Q: Does the NVIDIA GeForce GTX 670MX have any benchmark advantage?

A: Yes, the GTX 670MX has a recorded Geekbench Vulkan score of 5316. The FirePro W600 has no Vulkan benchmark score in the database, so the GTX 670MX is the only card with demonstrated Vulkan performance.

Q: How do the two cards compare in average benchmark score?

A: The FirePro W600 averages 6223 across its single benchmark. The GTX 670MX averages 5721 across its two benchmarks (OpenCL and Vulkan). This represents a difference of approximately 8.8%.

Q: What is the memory clock difference between the two?

A: The FirePro W600 runs its GDDR5 memory at 1000 MHz (4 Gbps effective). The GTX 670MX runs its GDDR5 memory at 700 MHz (2.8 Gbps effective).

Q: Which card has more shading units?

A: The NVIDIA GeForce GTX 670MX has 960 shading units. The AMD FirePro W600 has 512 shading units. The GTX 670MX also has 80 TMUs versus 32, and 24 ROPs versus 16.

Q: Are both cards the same age?

A: No. The FirePro W600 has a release date of 2012-06-12, while the GTX 670MX has a release date of 2012-09-30. The GTX 670MX launched about three and a half months later.

Architecture Differences

The AMD FirePro W600 uses the Cape Verde chip built on GCN 1.0 architecture, while the NVIDIA GeForce GTX 670MX uses the GK104 chip built on Kepler. Both are manufactured on a 28 nm process at TSMC, but the transistor counts differ substantially. The FirePro W600 packs 1,500 million transistors into a 123 mm² die, yielding a density of 12.2M transistors per mm². The GTX 670MX packs 3,540 million transistors into a 294 mm² die, yielding a slightly lower density of 12.0M per mm².

The FirePro W600 belongs to the FirePro GCN (Wx000) generation and succeeds the FirePro Terascale line. Its successor is the Radeon Pro Polaris. The GTX 670MX belongs to the GeForce 600M generation, succeeding the GeForce 500M and leading to the GeForce 700M. These lineage differences reflect distinct market positioning: the FirePro targets professional workstations, while the GTX 670MX targets mobile gaming laptops.

The shading architectures diverge significantly. The FirePro W600 has 512 shading units, 32 TMUs, and 16 ROPs. The GTX 670MX has 960 shading units, 80 TMUs, and 24 ROPs. The GTX 670MX's texture rate is exactly double at 48.08 GTexel/s versus 24.00 GTexel/s. The pixel rates are nearly identical: 12.00 GPixel/s for the FirePro W600 and 12.02 GPixel/s for the GTX 670MX.

FP32 throughput differs notably: the FirePro W600 delivers 768.0 GFLOPS, while the GTX 670MX delivers 1,153.9 GFLOPS. Neither card has recorded FP16, RT core, or tensor core data. Both support DirectX 12 (with the FirePro W600 at 11_1 and the GTX 670MX at 11_0), OpenGL 4.6, and Vulkan (1.2.170 for the FirePro, 1.2.175 for the GTX 670MX).

The FirePro W600 displays through 6x mini-DisplayPort 1.2 outputs, while the GTX 670MX's display outputs are listed as portable device dependent, reflecting its mobile nature. The FirePro W600 is single-slot with no power connectors and a suggested PSU of 250 W. The GTX 670MX has no slot width data, no power connector data, and no suggested PSU data, consistent with its likely soldered mobile implementation.

Specification Differences

The memory configurations present clear differences. The FirePro W600 has 2 GB of GDDR5 on a 128-bit bus, delivering 64.00 GB/s bandwidth. The GTX 670MX has 3 GB of GDDR5 on a 192-bit bus, delivering 67.20 GB/s bandwidth. The FirePro W600's memory clock is 1000 MHz (4 Gbps effective), while the GTX 670MX's is 700 MHz (2.8 Gbps effective). The larger bus on the GTX 670MX nearly compensates for its lower clock, resulting in only a 5% bandwidth advantage.

The GTX 670MX has a base clock of 601 MHz and a boost clock of 601 MHz, indicating a fixed clock. The FirePro W600 has no recorded base or boost clock, only its memory clock. This absence of clock data for the FirePro W600 makes direct frequency comparisons impossible.

Physical dimensions differ: the FirePro W600 measures 168 mm in length, 111 mm in height, and 20 mm in width. The GTX 670MX has no recorded dimensions, reflecting its mobile form factor where the card is integrated into a laptop chassis. The FirePro W600's single-slot design and lack of power connectors make it a low-profile installation, while the GTX 670MX's power requirements are unknown.

The launch MSRP for the FirePro W600 is 599 USD. The GTX 670MX has no launch MSRP recorded, consistent with mobile GPUs that are typically sold as part of complete laptops rather than as standalone products. The FirePro W600's production status is end-of-life, as is the GTX 670MX's. The FirePro W600's release date is 2012-06-12, and the GTX 670MX's is 2012-09-30.

The bus interface is identical: PCIe 3.0 x16 for both. The TDP is also identical at 75 W for each. The FirePro W600 is made by AMD, the GTX 670MX by NVIDIA, and both use TSMC as the foundry. The GTX 670MX has a larger die (294 mm² vs. 123 mm²) and more transistors (3,540 million vs. 1,500 million), but the FirePro W600 achieves a slightly higher transistor density (12.2M vs. 12.0M per mm²).

DETAILED SPECIFICATIONS

SPECIFICATION
FirePro W600
GTX 670MX
Core Specs
Shading Units
512
960 +87.5%
Shaders
512
960 +87.5%
TMUs
32
80 +150.0%
ROPs
16
24 +50.0%
Compute Units
8
Clocks
Base Clock
601 MHz
Boost Clock
601 MHz
GPU Clock
750 MHz
Memory Clock
1000 MHz 4 Gbps effective
700 MHz 2.8 Gbps effective
Memory
Memory Size
2 GB
3 GB
VRAM (MB)
2,048
3,072 +50.0%
Memory Type
GDDR5
GDDR5
Memory Bus
128 bit
192 bit
Bandwidth
64.00 GB/s
67.20 GB/s
Cache
L1 Cache
16 KB (per CU)
16 KB (per SMX)
L2 Cache
256 KB
384 KB
Performance
Pixel Rate
12.00 GPixel/s
12.02 GPixel/s
Texture Rate
24.00 GTexel/s
48.08 GTexel/s
FP32 (TFLOPS)
768.0 GFLOPS
1,153.9 GFLOPS
FP64 (TFLOPS)
48.00 GFLOPS (1:16)
48.08 GFLOPS (1:24)
Power
TDP
75 W
75 W
TDP (W)
75
75 0.0%
Suggested PSU
250 W
Power Connectors
None
None
Architecture
Architecture
GCN 1.0
Kepler
GPU Name
Cape Verde
GK104
Generation
FirePro GCN (Wx000)
GeForce 600M
Process Size
28 nm
28 nm
Transistors
1,500 million
3,540 million
Die Size
123 mm²
294 mm²
Foundry
TSMC
TSMC
Density
12.2M / mm²
12.0M / mm²
API Support
DirectX
12 (11_1)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.2.170
1.2.175
OpenCL
2.1 (1.2)
3.0
CUDA
3.0
Shader Model
6.5 (5.1)
6.5 (5.1)
Physical
Slot Width
Single-slot
Length
168 mm 6.6 inches
Height
111 mm 4.4 inches
Outputs
6x mini-DisplayPort 1.2
Portable Device Dependent
Bus Interface
PCIe 3.0 x16
PCIe 3.0 x16
Other
Launch Price
599 USD
Production
End-of-life
End-of-life
Predecessor
FirePro Terascale
GeForce 500M
Successor
Radeon Pro Polaris
GeForce 700M
View FirePro W600 Details View GeForce GTX 670MX Details