AMD Ryzen 5 Pro 3500U vs Intel Core i5-1035G4: What is the difference?
49points
AMD Ryzen 5 Pro 3500U
45points
Intel Core i5-1035G4
€218
Comparison winner
€218
vs
62 facts in comparison
AMD Ryzen 5 Pro 3500U
Intel Core i5-1035G4
Why is AMD Ryzen 5 Pro 3500U better than Intel Core i5-1035G4?
- 1.91x faster CPU speed?
4 x 2.1GHzvs4 x 1.1GHz - 900MHz faster GPU clock speed?
1200MHzvs300MHz - 64KB bigger L1 cache?
384KBvs320KB - Supports ECC memory?
- 1 more displays supported?
4vs3
Why is Intel Core i5-1035G4 better than AMD Ryzen 5 Pro 3500U?
- 1333MHz higher ram speed?
3733MHzvs2400MHz - 5°C higher maximum operating temperature?
100°Cvs95°C - 2nm smaller semiconductor size?
10nmvs12nm - 17.
1% higher PassMark result?
8196vs6999 - 2MB bigger L3 cache?
6MBvs4MB - 22.54GB/s more memory bandwidth?
58.3GB/svs35.76GB/s - 18.75% higher PassMark result (single)?
2337vs1968 - 32GB larger maximum memory amount?
64GBvs32GB
AMD Ryzen 5 5500U
vs
Intel Core i5-1135G7
AMD Ryzen 5 5600H
vs
Intel Core i5-12500H
Intel Core i5-1135G7
vs
Intel Core i5-1235U
Intel Core i3-1215U
vs
Intel Core i5-1135G7
AMD Ryzen 5 5600H
vs
Intel Core i5-11300H
AMD Ryzen 7 6800H
vs
Intel Core i7-12700H
AMD Ryzen 5 5600G
vs
AMD Ryzen 7 5700G
AMD Ryzen 5 5500U
vs
Intel Core i3-1115G4
AMD Ryzen 3 5300U
vs
Intel Core i3-1115G4
AMD Ryzen 3 3250U
vs
Intel Core i3-1115G4
Price comparison
Intel Core i5-1035G4
| Product | Store | Price | |
|---|---|---|---|
Intel Core i5-13400F Processor 20 MB Sma. ..Intel Core i5-13400F Processor 20 MB Smart Cache Box |
€218 | ||
| Intel S1700 Core i5 13400F Tray GEN13 | €218 | ||
| Intel Core i5-11400F 11. Generation Desk…Intel Core i5-11400F 11. Generation Desktop Prozessor (Basistakt: 2.6GHz Tuboboost: 4.4GHz, 6 Kerne, LGA1200) BX8070811400F | €130 | ||
| Intel S1700 Core i5 13600 Tray GEN13 | €298 | ||
User reviews
Performance
CPU speed
4 x 2.1GHz
4 x 1.1GHz
The CPU speed indicates how many processing cycles per second can be executed by a CPU, considering all of its cores (processing units). It is calculated by adding the clock rates of each core or, in the case of multi-core processors employing different microarchitectures, of each group of cores.
CPU threads
More threads result in faster performance and better multitasking.
turbo clock speed
3.7GHz
3.
7GHz
When the CPU is running below its limitations, it can boost to a higher clock speed in order to give increased performance.
Has an unlocked multiplier
✖AMD Ryzen 5 Pro 3500U
✖Intel Core i5-1035G4
Some processors come with an unlocked multiplier which makes them easy to overclock, allowing you to gain increased performance in games and other apps.
L2 cache
A larger L2 cache results in faster CPU and system-wide performance.
L3 cache
A larger L3 cache results in faster CPU and system-wide performance.
L1 cache
A larger L1 cache results in faster CPU and system-wide performance.
L2 core
0.5MB/core
0.5MB/core
More data can be stored in the L2 cache for access by each core of the CPU.
L3 core
1MB/core
1.5MB/core
More data can be stored in the L3 cache for access by each core of the CPU.
Benchmarks
PassMark result
This benchmark measures the performance of the CPU using multiple threads.
PassMark result (single)
This benchmark measures the performance of the CPU using a single thread.
Cinebench R20 (multi) result
Cinebench R20 is a benchmark tool that measures a CPU’s multi-core performance by rendering a 3D scene.
Cinebench R20 (single) result
Cinebench R20 is a benchmark tool that measures a CPU’s single-core performance by rendering a 3D scene.
Geekbench 5 result (multi)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
Geekbench 5 is a cross-platform benchmark that measures a processor’s multi-core performance. (Source: Primate Labs, 2023)
Geekbench 5 result (single)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
Geekbench 5 is a cross-platform benchmark that measures a processor’s single-core performance. (Source: Primate Labs, 2023)
Blender (bmw27) result
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
The Blender (bmw27) benchmark measures the performance of a processor by rendering a 3D scene.
More powerful processors can render the scene in less time.
Blender (classroom) result
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
The Blender (classroom) benchmark measures the performance of a processor by rendering a 3D scene. More powerful processors can render the scene in less time.
performance per watt
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
This means the CPU is more efficient, giving a greater amount of performance for each watt of power used.
Integrated graphics
GPU clock speed
1200MHz
300MHz
The graphics processing unit (GPU) has a higher clock speed.
GPU turbo
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
1050MHz
When the GPU is running below its limitations, it can boost to a higher clock speed in order to give increased performance.
GPU execution units
Unknown.
Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
A graphics processing unit (GPU) with a greater number of execution units can deliver better graphics.
supported displays
Using multiple displays you can create a larger workspace, making it easier to work across multiple apps.
DirectX version
DirectX is used in games, with newer versions supporting better graphics.
OpenGL version
OpenGL is used in games, with newer versions supporting better graphics.
OpenCL version
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
Some apps use OpenCL to apply the power of the graphics processing unit (GPU) for non-graphical computing. Newer versions introduce more functionality and better performance.
texture mapping units (TMUs)
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
TMUs take textures and map them to the geometry of a 3D scene.
More TMUs will typically mean that texture information is processed faster.
render output units (ROPs)
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
The ROPs are responsible for some of the final steps of the rendering process, writing the final pixel data to memory and carrying out other tasks such as anti-aliasing to improve the look of graphics.
Memory
RAM speed
2400MHz
3733MHz
It can support faster memory, which will give quicker system performance.
maximum memory bandwidth
35.76GB/s
58.3GB/s
This is the maximum rate that data can be read from or stored into memory.
DDR memory version
DDR (Double Data Rate) memory is the most common type of RAM. Newer versions of DDR memory support higher maximum speeds and are more energy-efficient.
memory channels
More memory channels increases the speed of data transfer between the memory and the CPU.
maximum memory amount
The maximum amount of memory (RAM) supported.
bus transfer rate
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
The bus is responsible for transferring data between different components of a computer or device.
Supports ECC memory
✔AMD Ryzen 5 Pro 3500U
✖Intel Core i5-1035G4
Error-correcting code memory can detect and correct data corruption. It is used when is it essential to avoid corruption, such as scientific computing or when running a server.
eMMC version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
A higher version of eMMC allows faster memory interfaces, having a positive effect on the performance of a device. For example, when transferring files from your computer to the internal storage over USB.
bus speed
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown.
Help us by suggesting a value. (Intel Core i5-1035G4)
The bus is responsible for transferring data between different components of a computer or device.
Features
instruction sets
SSE 4.2, SSE 4.1, AVX, AES, FMA3, F16C, MMX
SSE 4.2, SSE 4.1, AVX, AES, FMA3, F16C, MMX
Instruction sets are sets of codes that the CPU runs for certain functions.
uses multithreading
✔AMD Ryzen 5 Pro 3500U
✔Intel Core i5-1035G4
Multithreading technology (such as Intel’s Hyperthreading or AMD’s Simultaneous Multithreading) provides increased performance by splitting each of the processor’s physical cores into virtual cores, also known as threads. This way, each core can run two instruction streams at once.
bits executed at a time
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
NEON provides acceleration for media processing, such as listening to MP3s.
Has TrustZone
✖AMD Ryzen 5 Pro 3500U
✖Intel Core i5-1035G4
A technology integrated into the processor to secure the device for use with features such as mobile payments and streaming video using digital rights management (DRM).
front-end width
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
The CPU can decode more instructions per clock (IPC), meaning that the CPU performs better
Has NX bit
✔AMD Ryzen 5 Pro 3500U
✔Intel Core i5-1035G4
NX bit helps protect the computer from malicious attacks.
VFP version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 3500U)
Unknown. Help us by suggesting a value. (Intel Core i5-1035G4)
Vector Floating-Point (VFP) is used by the processor to deliver increased performance in areas such as digital imaging.
Miscellaneous
Has AES
✔AMD Ryzen 5 Pro 3500U
✔Intel Core i5-1035G4
AES is used to speed up encryption and decryption.
Has integrated LTE
✖AMD Ryzen 5 Pro 3500U
✖Intel Core i5-1035G4
The system on a chip (SoC) has an integrated LTE cellular chip. LTE is capable of downloading at faster speeds than older, 3G technology.
Price comparison
Intel Core i5-1035G4
| Product | Store | Price | |
|---|---|---|---|
| Intel Core i5-13400F Processor 20 MB Sma…Intel Core i5-13400F Processor 20 MB Smart Cache Box | €218 | ||
| Intel S1700 Core i5 13400F Tray GEN13 | €218 | ||
| Intel Core i5-11400F 11. Generation Desk…Intel Core i5-11400F 11. Generation Desktop Prozessor (Basistakt: 2.6GHz Tuboboost: 4.4GHz, 6 Kerne, LGA1200) BX8070811400F | €130 | ||
| Intel S1700 Core i5 13600 Tray GEN13 | €298 | ||
AMD Ryzen 5 Pro 4650U vs Intel Core i5-10210U: What is the difference?
Search
smartphonesgraphics cardswireless earbudsCPUs
45points
AMD Ryzen 5 Pro 4650U
46points
Intel Core i5-10210U
€149
€127
€149
Comparison winner
€127
vs
61 facts in comparison
AMD Ryzen 5 Pro 4650U
Intel Core i5-10210U
Why is AMD Ryzen 5 Pro 4650U better than Intel Core i5-10210U?
- 1.
97x faster CPU speed?
6 x 2.1GHzvs4 x 1.6GHz - 1600MHz higher ram speed?
4266MHzvs2666MHz - 4 more CPU threads?
12vs8 - 7nm smaller semiconductor size?
7nmvs14nm - 2MB bigger L2 cache?
3MBvs1MB - 1.99x higher PassMark result?
12839vs6464 - 2MB bigger L3 cache?
8MBvs6MB - 128KB bigger L1 cache?
384KBvs256KB
Why is Intel Core i5-10210U better than AMD Ryzen 5 Pro 4650U?
- 10W lower TDP?
15Wvs25W - 3.41x higher Cinebench R20 (multi) result?
1797vs527 - 1.51x higher Cinebench R20 (single) result?
426vs282 - 0.17MB/core more L3 cache per core?
1.5MB/corevs1.33MB/core
AMD Ryzen 5 Pro 4650U
vs
Intel Core i5-1135G7
Intel Core i5-10210U
vs
AMD Ryzen 7 3700U
AMD Ryzen 5 Pro 4650U
vs
Intel Core i7-10510U
Intel Core i5-10210U
vs
AMD Ryzen 5 3500U
AMD Ryzen 5 Pro 4650U
vs
AMD Ryzen 5 5500U
Intel Core i5-10210U
vs
Intel Core i3-1115G4
AMD Ryzen 5 Pro 4650U
vs
Intel Core i7-1065G7
Intel Core i5-10210U
vs
Intel Core i7-8550U
AMD Ryzen 5 Pro 4650U
vs
Intel Core i7-8550U
Intel Core i5-10210U
vs
AMD Ryzen 5 5500U
AMD Ryzen 5 Pro 4650U
vs
AMD Ryzen 5 Pro 5650U
Intel Core i5-10210U
vs
Intel Core i5-1135G7
AMD Ryzen 5 Pro 4650U
vs
Intel Core i5-8265U
Intel Core i5-10210U
vs
Intel Core i3-10110U
AMD Ryzen 5 Pro 4650U
vs
Intel Core i5-8250U
Intel Core i5-10210U
vs
Intel Core i5-8250U
AMD Ryzen 5 Pro 4650U
vs
Intel Core i5-10300H
Intel Core i5-10210U
vs
AMD Ryzen 5 3450U
Intel Core i5-10210U
vs
Intel Core i7-10510U
Price comparison
AMD Ryzen 5 Pro 4650U
| Product | Store | Price | |
|---|---|---|---|
| AMD Ryzen 5 5600X Box, Large | €149 | ||
AMD Ryzen 7 5700X Prozessor (Basistakt 3. ..AMD Ryzen 7 5700X Prozessor (Basistakt 3.4GHz, Max. Leistungstakt bis zu 4.6GHz, 8 Kerne, L3-Cache 32MB, Socket AM4, ohne Kühler) 100-100000926WOF, Schwarz |
€184 | ||
| AMD Ryzen 5 5600G mit AMD Radeon Grafik …AMD Ryzen 5 5600G mit AMD Radeon Grafik (6x 3,9 GHz) 19MB Sockel AM4 CPU BOX | €123 | ||
| AMD Ryzen 5 Pro 4650G Prozessor 3,7 GHz …AMD Ryzen 5 Pro 4650G Prozessor 3,7 GHz 8 MB L3 | €119 | ||
Intel Core i5-10210U
| Product | Store | Price | |
|---|---|---|---|
| Intel Core i5-11400F 11. Generation Desk…Intel Core i5-11400F 11. Generation Desktop Prozessor (Basistakt: 2.6GHz Tuboboost: 4.4GHz, 6 Kerne, LGA1200) BX8070811400F | €127 | ||
| Intel Core i5-13400F Processor 20 MB Sma…Intel Core i5-13400F Processor 20 MB Smart Cache Box | €218 | ||
Intel Core i5-13400 Processor 20 MB Smar. ..Intel Core i5-13400 Processor 20 MB Smart Cache Box |
€219 | ||
| Intel Core i5-12400F 12. Generation Desk…Intel Core i5-12400F 12. Generation Desktop Prozessor (Basistakt: 2.5GHz, 6 Kerne, LGA1700, RAM DDR4 und DDR5 bis zu 128GB) BX8071512400F | €149 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €362 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €371 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €371 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €375 | ||
User reviews
Overall rating
AMD Ryzen 5 Pro 4650U
0 User reviews
AMD Ryzen 5 Pro 4650U
0.
0/10
0 User reviews
Intel Core i5-10210U
5 User reviews
Intel Core i5-10210U
6.4/10
5 User reviews
Features
Value for money
No reviews yet
6.2/10
5 votes
Gaming
No reviews yet
5.6/10
5 votes
Performance
No reviews yet
6.0/10
5 votes
Reliability
No reviews yet
7.0/10
5 votes
Energy efficiency
No reviews yet
6.8/10
5 votes
Performance
CPU speed
6 x 2.1GHz
4 x 1.6GHz
The CPU speed indicates how many processing cycles per second can be executed by a CPU, considering all of its cores (processing units). It is calculated by adding the clock rates of each core or, in the case of multi-core processors employing different microarchitectures, of each group of cores.
CPU threads
More threads result in faster performance and better multitasking.
turbo clock speed
4.2GHz
When the CPU is running below its limitations, it can boost to a higher clock speed in order to give increased performance.
Has an unlocked multiplier
✖AMD Ryzen 5 Pro 4650U
✖Intel Core i5-10210U
Some processors come with an unlocked multiplier which makes them easy to overclock, allowing you to gain increased performance in games and other apps.
L2 cache
A larger L2 cache results in faster CPU and system-wide performance.
L3 cache
A larger L3 cache results in faster CPU and system-wide performance.
L1 cache
A larger L1 cache results in faster CPU and system-wide performance.
L2 core
0.5MB/core
0.25MB/core
More data can be stored in the L2 cache for access by each core of the CPU.
L3 core
1.33MB/core
1.5MB/core
More data can be stored in the L3 cache for access by each core of the CPU.
Benchmarks
PassMark result
This benchmark measures the performance of the CPU using multiple threads.
PassMark result (single)
This benchmark measures the performance of the CPU using a single thread.
Cinebench R20 (multi) result
Cinebench R20 is a benchmark tool that measures a CPU’s multi-core performance by rendering a 3D scene.
Cinebench R20 (single) result
Cinebench R20 is a benchmark tool that measures a CPU’s single-core performance by rendering a 3D scene.
Geekbench 5 result (multi)
Geekbench 5 is a cross-platform benchmark that measures a processor’s multi-core performance. (Source: Primate Labs, 2023)
Geekbench 5 result (single)
Geekbench 5 is a cross-platform benchmark that measures a processor’s single-core performance. (Source: Primate Labs, 2023)
Blender (bmw27) result
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
790.56seconds
The Blender (bmw27) benchmark measures the performance of a processor by rendering a 3D scene. More powerful processors can render the scene in less time.
Blender (classroom) result
Unknown.
Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
2274.84seconds
The Blender (classroom) benchmark measures the performance of a processor by rendering a 3D scene. More powerful processors can render the scene in less time.
performance per watt
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
This means the CPU is more efficient, giving a greater amount of performance for each watt of power used.
Integrated graphics
GPU clock speed
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
300MHz
The graphics processing unit (GPU) has a higher clock speed.
GPU turbo
1500MHz
1100MHz
When the GPU is running below its limitations, it can boost to a higher clock speed in order to give increased performance.
GPU execution units
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
A graphics processing unit (GPU) with a greater number of execution units can deliver better graphics.
supported displays
Using multiple displays you can create a larger workspace, making it easier to work across multiple apps.
DirectX version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
DirectX is used in games, with newer versions supporting better graphics.
OpenGL version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
OpenGL is used in games, with newer versions supporting better graphics.
OpenCL version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
Some apps use OpenCL to apply the power of the graphics processing unit (GPU) for non-graphical computing. Newer versions introduce more functionality and better performance.
texture mapping units (TMUs)
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
TMUs take textures and map them to the geometry of a 3D scene.
More TMUs will typically mean that texture information is processed faster.
render output units (ROPs)
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
The ROPs are responsible for some of the final steps of the rendering process, writing the final pixel data to memory and carrying out other tasks such as anti-aliasing to improve the look of graphics.
Memory
RAM speed
4266MHz
2666MHz
It can support faster memory, which will give quicker system performance.
maximum memory bandwidth
68.27GB/s
41.66GB/s
This is the maximum rate that data can be read from or stored into memory.
DDR memory version
DDR (Double Data Rate) memory is the most common type of RAM. Newer versions of DDR memory support higher maximum speeds and are more energy-efficient.
memory channels
More memory channels increases the speed of data transfer between the memory and the CPU.
maximum memory amount
The maximum amount of memory (RAM) supported.
bus transfer rate
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
The bus is responsible for transferring data between different components of a computer or device.
Supports ECC memory
✖AMD Ryzen 5 Pro 4650U
✖Intel Core i5-10210U
Error-correcting code memory can detect and correct data corruption. It is used when is it essential to avoid corruption, such as scientific computing or when running a server.
eMMC version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
A higher version of eMMC allows faster memory interfaces, having a positive effect on the performance of a device. For example, when transferring files from your computer to the internal storage over USB.
bus speed
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown.
Help us by suggesting a value. (Intel Core i5-10210U)
The bus is responsible for transferring data between different components of a computer or device.
Features
instruction sets
SSE 4.2, SSE 4.1, AVX, AES, FMA3, F16C, MMX
SSE 4.2, SSE 4.1, AVX, AES, FMA3, F16C, MMX
Instruction sets are sets of codes that the CPU runs for certain functions.
uses multithreading
✔AMD Ryzen 5 Pro 4650U
✔Intel Core i5-10210U
Multithreading technology (such as Intel’s Hyperthreading or AMD’s Simultaneous Multithreading) provides increased performance by splitting each of the processor’s physical cores into virtual cores, also known as threads. This way, each core can run two instruction streams at once.
bits executed at a time
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
NEON provides acceleration for media processing, such as listening to MP3s.
Has TrustZone
✖AMD Ryzen 5 Pro 4650U
✖Intel Core i5-10210U
A technology integrated into the processor to secure the device for use with features such as mobile payments and streaming video using digital rights management (DRM).
front-end width
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
The CPU can decode more instructions per clock (IPC), meaning that the CPU performs better
Has NX bit
✔AMD Ryzen 5 Pro 4650U
✔Intel Core i5-10210U
NX bit helps protect the computer from malicious attacks.
VFP version
Unknown. Help us by suggesting a value. (AMD Ryzen 5 Pro 4650U)
Unknown. Help us by suggesting a value. (Intel Core i5-10210U)
Vector Floating-Point (VFP) is used by the processor to deliver increased performance in areas such as digital imaging.
Miscellaneous
Has AES
✔AMD Ryzen 5 Pro 4650U
✔Intel Core i5-10210U
AES is used to speed up encryption and decryption.
Price comparison
AMD Ryzen 5 Pro 4650U
| Product | Store | Price | |
|---|---|---|---|
| AMD Ryzen 5 5600X Box, Large | €149 | ||
AMD Ryzen 7 5700X Prozessor (Basistakt 3. ..AMD Ryzen 7 5700X Prozessor (Basistakt 3.4GHz, Max. Leistungstakt bis zu 4.6GHz, 8 Kerne, L3-Cache 32MB, Socket AM4, ohne Kühler) 100-100000926WOF, Schwarz |
€184 | ||
| AMD Ryzen 5 5600G mit AMD Radeon Grafik …AMD Ryzen 5 5600G mit AMD Radeon Grafik (6x 3,9 GHz) 19MB Sockel AM4 CPU BOX | €123 | ||
| AMD Ryzen 5 Pro 4650G Prozessor 3,7 GHz …AMD Ryzen 5 Pro 4650G Prozessor 3,7 GHz 8 MB L3 | €119 | ||
Intel Core i5-10210U
| Product | Store | Price | |
|---|---|---|---|
| Intel Core i5-11400F 11. Generation Desk…Intel Core i5-11400F 11. Generation Desktop Prozessor (Basistakt: 2.6GHz Tuboboost: 4.4GHz, 6 Kerne, LGA1200) BX8070811400F | €127 | ||
| Intel Core i5-13400F Processor 20 MB Sma…Intel Core i5-13400F Processor 20 MB Smart Cache Box | €218 | ||
Intel Core i5-13400 Processor 20 MB Smar. ..Intel Core i5-13400 Processor 20 MB Smart Cache Box |
€219 | ||
| Intel Core i5-12400F 12. Generation Desk…Intel Core i5-12400F 12. Generation Desktop Prozessor (Basistakt: 2.5GHz, 6 Kerne, LGA1700, RAM DDR4 und DDR5 bis zu 128GB) BX8071512400F | €149 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €362 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €371 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €371 | ||
| Mini-PC/Barebone Komplettsystem Intel Co…Mini-PC/Barebone Komplettsystem Intel Core i5-10210U, 8GB RAM (IGP/diverse Grafikkarten) | €375 | ||
This page is currently only available in English.
AMD Ryzen 5 3400G and Intel Core i5-9600K with integrated and discrete graphics in our test methodology
Over the past three years, the bulk of our testing of processors and «finished» computers has been on systems with discrete graphics. First of all, this is due to the fact that full-fledged competition in the field of high performance has resumed, but many of the corresponding solutions do not have an integrated GPU at all. AMD, for example, has a traditional approach and has not changed for almost 10 years: pure processors are nominally more powerful than APUs (as the company calls its integrated solutions) at least twice. In Intel, in conditions of shortage, they returned to the practice of releasing mass Cores with a blocked video core. As for HEDT, both manufacturers have the same approach here: only discrete.
The second reason is somewhat archaic modern integrashki. Intel hasn’t changed anything since 2017 — the UHD Graphics 600th line started with Kaby Lake, «crawled» into Coffee Lake in two versions, and today they are used in Comet Lake.
By the way, by and large, they have no serious differences from the 500th line, and this is the Skylake of 2015. AMD had a major change in 2018, but the integrated Vega, although more productive than Intel solutions, often does not allow you to play modern games even “at the minimum”. In general, it is not possible to seriously consider integration programs. Yes, most computers do without discrete graphics cards, but only because of approach «works — and okay» . This approach does not suit the most demanding users, so it has long been considered that «without a video card — nowhere.»
Returning to our tests, in the case of desktop systems, we can easily test all processors with the same video card. For many years, this was our standard practice, completely leveling the influence of a video card on performance (more precisely, bringing it to a common denominator). But this practice is not always applicable. For example, in a compact system, the choice of graphic solutions is usually limited: even if there is a discrete one, it is usually not at all what we use in a normal situation.
Or the budget segment, where it is not uncommon to save money on a video card (at least at the first stage). Therefore, as the number of tested configurations increases, it is time to address this issue as well.
Testers
We took a couple of processors — AMD Ryzen 5 3400G and Intel Core i5-9600K. Why exactly them? The second, for example, is used by us as a scale unit — that is, the performance of a system with this processor is taken as 100 points. And until recently, the first one remained the most powerful AMD solution with integrated graphics, and although it was still inferior to the same i5-9600K in terms of the processor component, it significantly outperformed it in terms of the graphics component. This pair is enough for us to solve the set tasks — full equalization of the processors is not required. It is important that there are two different ones, and the models of both manufacturers are taken and with their own graphic solutions of both companies. Intel has been absent from the discrete graphics market for a long time and is only going to return to it.
And AMD uses the Vega 11 core, which is architecturally similar to that used in our standard Vega 56. Of course, it makes no sense to compare these integrated chips in terms of performance in gaming applications, but we won’t need games today.
What else do you need? In the GPU market, unlike x86 processors, there are still three players, not two. So we will also add a video card based on Nvidia GeForce GTX 1070. In terms of performance, this solution is comparable to Vega 56 and is radically different from any integrated solutions. Functionally, it is really a «third force» with its own architecture and its own drivers. Without her, the comparison would be incomplete. With her … also not completely complete, since both AMD and Nvidia have a lot of discrete solutions — mostly slower than this pair, but there are also faster ones. But we are more important than trends today. In other words, where there is no effect of a video card on performance, there will be no difference between different video cards.
And if suddenly there is a difference, this is already a reason for more subtle research.
Other environment — traditionally the same and affecting the results in the same way. The main thing is six pairs of «processor + video card», which we will compare with each other. More precisely, we will compare two triplets: each processor with its own integrated GPU and with two discrete ones. We already know that with the same graphics card, the performance of the Ryzen 5 3400G is 72.5% that of the Core i5-9600K. But we still don’t know how this ratio will change with different video cards. And it should.
Testing goals
It should be noted right away that the purpose of this material is not a full comparison of different graphic solutions in various tasks with different data. It’s time to do this too, but this process is not instantaneous — and it won’t fit in one material.
Our task is simpler and more practical. There is a set of programs that we use when testing processors and systems.
There is a data set with which they operate in tests. There are settings that are the same in all tests. And, by the way, including the use of the GPU to help the CPU, where possible. Without special “shifting” the load with additional settings and / or presets. That is, in simple words, we did not turn off anything, but we did not turn it on on purpose either. In fact, the program settings for 99% are left at default values: how it is installed and how it works.
Can we see how the results are affected by changing the video card? Can. Need to do it? Need to. At least within the framework of the workflow — to understand whether it is possible to compare the results of different processors with different graphic solutions. Since if it suddenly turns out (actually not) that the influence of the video card can be such that, for example, swapping Ryzen 5 3400G and Core i5-9600K, then we can correctly compare processors only when using the same video card. If the influence of the video card turns out to be zero, then we can correctly compare in any conditions.
If not zero, but limited, then it is possible to compare, but up to this influence.
In addition, we measure not only performance but also energy consumption. Moreover, we measure in the only correct way (unfortunately, rarely encountered): simultaneously. Thus, the result of our tests is not only time consumption , but also energy consumption . Therefore, it is imperative to look at how the video card affects this. In the end, the legend that a discrete video card reduces the power consumption of the processor is still alive in some places.
Testing methodology
Testing methodology is described in detail in a separate article, and the results of all tests are available in a separate table in Microsoft Excel format (for this article — a separate one: due to its specificity). Directly in the articles, we use the processed results: normalized with respect to the reference system (Intel Core i5-9600K with 16 GB of memory, AMD Radeon Vega 56 video card and SATA SSD — she is also directly involved in today’s article) and grouped by areas of application of the computer.
Accordingly, all diagrams related to applications have dimensionless scores, so more is always better. And today we don’t need game tests even in a rudimentary form — since the influence of the GPU on them, in general, does not even make sense to discuss 🙂
iXBT Application Benchmark 2020
You can catch decimal fleas, but it’s clear that there is no direct influence of the GPU on the results as such. Some kind of indirect when interacting with different drivers is possible, but still within the margin of error. Therefore, although the difference remains stable when repeating the tests in any sequence, you should not pay attention to it.
Similar to the previous case. Although GPU computing can also be used for 3D rendering, this has to be done on purpose and prepared. Nothing happens automatically in the programs we use.
Now that’s interesting. True, the interaction of programs exclusively with Intel graphic solutions, which so far exist only in the form integrated into the processor, has some peculiarities.
And in order to deal with the issue thoroughly, it makes sense to look at the detailed results of the programs included in this group.
| Video content creation | IGP | Vega 56 | GTX 1070 |
|---|---|---|---|
| Adobe Premiere Pro CC 2019v13.01.13, c | 158.08 | 298.90 | 329.60 |
| Magix Vegas Pro 16.0, c | 353.33 | 363.50 | 361.00 |
| Magix Movie Edit Pro 2019 Premium v18.03.261, c | 65.73 | 413.34 | 449.64 |
| Adobe After Effects CC 2019 v16.0.1, since | 474.00 | 468.67 | 476.00 |
| Photodex ProShow Producer v9.0.3782, c | 183.78 | 191.12 | 176.52 |
In Vegas Pro and Photodex ProShow Producer, there is a slight scatter of results, which fits well into the general outline. Vegas «knows how» to use hardware decoders and encoders to the fullest, but it does this only when the appropriate presets are selected.
In tests, we «asked» him to rely on the processor — that’s how he behaves.
Premiere tries to use at least the hardware decoder by default. But the 2019 version we use only knows one of them — the Intel integrated graphics decoder. Its use removes a lot of work from the processor cores, so the performance is doubled. At the moment, Adobe is working on making everything work on different discrete and integrated solutions, and most likely, this work will soon be successfully completed. Why they started with Intel is also understandable: a huge number of laptops still use Intel processors, and almost all of them, even with a discrete video adapter, can be loaded with work and integrated. With our choice of testing tools, we just got to the moment when some GPUs are already in use (this was not the case before), but not all of them. And this factor should be taken into account. As well as the behavior of Movie Edit — which, having «discovered» the Intel GPU, strives to transfer all the work to it in general, so, for example, the processor cores in this mode are less than half loaded (in Premiere at least ⅔, and when using discrete graphics , which is «not supported» yet — at 100%).
This makes the results of different platforms not quite comparable — they differ by almost an order of magnitude. On the other hand… Everyone knows © that AMD APUs have significantly better graphics performance — this is clearly seen, for example, in games. And here is an example of the fact that graphics in Intel processors can be significantly more than compatible with with different software. And this is also quite an objective result.
Adobe After Effects stands apart. The results seem to be even… But in fact, the program makes great use of different GPUs — it just uses them equally efficiently. If we completely disable the use of the GPU by switching to Software Mode, then our test will take not 500, but more than 1100 seconds! A large part of this, however, falls on the conversion of the color space: without it «in the software» the test takes about 700 seconds. And “not in software” — 3-5. seconds. Total! Again — on any GPU. True, with a final warning that the result obtained may differ from the «reference» software.
With color conversion, there are no questions, but the effect of GPU acceleration is more modest. This is also, in general, a useful result: it shows that the idea of GPGPU computing should be treated with caution. They already work, they can be used — but it is not always easy to calculate the «economic effect», since it may depend not only on a specific program (and different versions of one program — this is in this issue different programs ), but also from a set of tasks, and even from data.
We return to the ground. At the moment, the situation is this, although a little earlier Phase One Capture One Pro (v10) behaved similarly to some video processing programs. Only he, on the contrary, did not perceive Intel HD Graphics as a video accelerator, but he used the Nvidia GeForce GTX 1070 as such, speeding up work at times. In the 12th version, the acceleration has not gone away, only now the program can equally use different GPUs. This should not be abandoned, since performance increases just as sharply, but on all modern graphic solutions it is almost the same.
In the other two programs of this group, the transition to purely programmatic methods also reduces productivity. But by default GPUs are used and different GPUs are used in the same way.
As expected, almost perfect similarity of results. In principle, this task is ideally parallelized, so, theoretically, it could also be “transferred” to the GPU. But in practice this is not necessary: the usual practice is to page-by-page recognition in interactive mode, which takes a little time — and only one processor core. And batch processing is applicable only in more specific cases, where, nevertheless, a modern multi-core processor is still sufficient to obtain acceptable performance. So, there is no point in complicating anything.
Ideas to implement GPGPU in the work of archivers have been around for a long time, special versions of programs with OpenCL support were also found on the market, but they turned out to be not very popular — in practice, even here, typical tasks are performed quickly enough to make you no longer want to change anything just for the sake of speed.
In this case, too, so far. Perhaps something will change, but hardly radically. Most likely, as in most cases, using a GPU can speed up programs, but the specific GPU will not be too important in this case. At times it comes to the ridiculous — when is less powerful than , but with more memory than is more powerful than with less memory. In games, this does not happen — unlike.
And in general, it is interesting that the «bang» of a pair of programs in one group significantly changed the state of affairs in it, but, being «smeared» into a large number of heterogeneous software, it almost disappeared. Although 7% is quite serious, since it exceeds at least the difference between the «neighboring» processors in the lines, so in the articles we will try not to get carried away with such comparisons if possible. But different discrete GPUs within the framework of the current version of the methodology do not affect performance — and this is good.
For the purity of the experiment, it’s better to compare processors under the same conditions, but if it doesn’t work out, just remember that AMD and Nvidia graphics solutions behave the same in the tasks we use, with the settings we use, and on our data sets.
Power Consumption and Energy Efficiency
Important note: the self-made meter we currently use does not work with additional «tails» for powering video cards. You can do it, but it wasn’t necessary. We register only that electricity that is supplied from the PSU to the board — through the main ATX connector and through 1-2 officially designed to power the processor. Such a clear separation is far from being for all platforms, and not only historical ones: modern AMD solutions also use a “hybrid” power scheme. Therefore, we work with the “total power”, which is almost complete for a system without discrete graphics (minus SATA drives), and with a video card, the main part of the electricity it consumes has long been supplied not through the PCIe slot.
We measure only what goes through this connector. Therefore, by the way, it makes no sense to become attached to the fact that the two platforms have different leaders in terms of gluttony: you also need to check how much each of the video cards consumes in total.
And what happens to the processor? For AM4, unfortunately, it is still impossible to determine this: the built-in sensors show (at least up to Zen + inclusive) the temperature in Karaganda, and both of the wires are actually used. At Intel, starting with LGA1150 (and in newer platforms), everything is smooth: processor power lines “hang” exclusively on an additional 12 V connector. And almost the same amount of energy “passes” through it — regardless of the video card used. Unless the maximum when the integrated video core is running is higher — just in the pair of programs where its operation leads to a significant increase in performance, so + 10% to power consumption is not a pity for this (especially since, as shown above, discrete GPUs «eat «much more, and not only in «their» power lines).
On average, the installation of a video card, contrary to one of the common urban legends, does not reduce the power consumption of the processor. Live with it 🙂
Since the performance with integrated graphics is not lower, and the power consumption of the platform is lower, the power efficiency also increases, including for AMD APUs, although the integrated Vega is similar in terms of performance impact to the discrete one. As a result, we remember that different systems should not be compared by this parameter: it will not work correctly.
Total
Both legends popular in narrow circles — «a discrete video card reduces the power consumption (and heat dissipation) of the processor» and «without a discrete video card in our time anywhere» — turned out to be untenable. The first one was completely and completely, but no one doubted this: the introduction of integrated graphics took place, among other things, to save energy. The second one is untenable in this formulation, since at least one class of software is known, in which the truth is “nowhere” — 3D games.