AMD’s Ryzen 7000 desktop processors will reportedly top 5.7 GHz in the case of the Zen giant’s top-of-the-line 7950X, when they launch later this quarter.
The industry watchers at Wccftech claim to have obtained detailed specs for AMD’s next-gen Zen 4 desktop CPUs, codenamed Raphael. If true, they offer some interesting insights into where AMD is going with its silicon designs.
It appears AMD is focusing more on IPC and clock gains in this generation and isn’t as interested in competing with Intel on core count. Here’s a breakdown of the SKUs per the leak:
- Ryzen 9 7950X: 16 core / 32 threads, with base clock of 4.5GHz, a boost clock of 5.7GHz, a TDP of 170W, and 64MB of L3 / 16MB of L2 cache.
- Ryzen 9 7900X: 12 core / 24 threads, with a base clock of 4.7GHz, a boost clock of 5.6GHz, a TDP of 170W, and 64MB of L3 / 12MB of L2 cache.
- Ryzen 7 7700X: 8 core / 16 threads, with a base of clock of 4.5GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 8MB of L2 cache.
- Ryzen 5 7600X: 6 core / 12 threads, with a base clock of 4.7GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 6MB of L2 cache.
All four SKUs will be fabbed using TSMC’s 5nm process. AMD confirmed earlier it will launch 5nm Zen 4 Ryzen 7000 parts this quarter.
TDP lifts clocks
If the leak is correct, the four chips see a notable increase in stated TDP over the 5000-series counterparts they replace. On the high-end, AMD 7000-series Ryzen 9 processors will reportedly suck back an additional 65W compared to the 5950X and 5900X. Meanwhile, AMD’s Ryzen 7 and Ryzen 5 processors have seen their rated TDPs boosted by roughly 40W.
It should be noted that a chip’s actual power consumption will likely be higher as AMD’s stated TDPs don’t actually reflect real-world power consumption. For example, under stock settings, AMD’s Ryzen 5900X routinely pulls between 120-140W of power under load, as long as the chip has adequate thermal headroom. The 5900X can suck down as much as 200W when the chipmaker’s Precision Boost Overdrive — an automated overlocking profile found in the bios — is enabled.
Unless AMD has changed the way it reports TDP since the launch of its 5000-series parts, there’s a good chance the chip designer’s Ryzen 7900 and 7950 parts could near 300W in real-world power consumption, putting them in the same territory as Intel’s 12th-gen chips.
- This credit card-sized PC board can use an Intel Core i7
- AMD boasts of record sales, says 5nm Zen 4 Ryzen 7000 coming this quarter
- Linus Torvalds releases Linux 5.19 – using Asahi on an Arm-powered Mac
- Intel experiences another kind of meltdown
However, the higher TDP appears to have netted AMD a substantial clock improvement over its previous-gen offerings. In fact, in a like-for-like comparison, the entire lineup boasts base clocks at 1 GHz or more over the chips they replace. In many cases, the processors’ base clocks now exceed their 5000-series equivalent’s boost clocks.
Cache clues
The leak also offers some insight into the chiplet architecture underpinning AMD’s Ryzen 7000 processors.
Starting off with the Ryzen 7600X, the package features 32MB of L3 cache — the same as the 5600X — but sees its L2 cache doubled to 6MB, or a full 1MB per core. By extension the 7700X features 8MB of L2.
Based on this, we can conclude that AMD is sticking with the dual chiplet architecture for its Ryzen 9 processors with either eight or six cores per core-complex die as we saw with previous generation Ryzen 5000 chips.
This isn’t terribly surprising as AMD hasn’t increased the core count with this generation of chips, at least not yet.
Will AMD nerf overclocking?
Finally, it’s unclear whether AMD will carry forward its longstanding tradition of unlocking every chip for the purposes of overclocking.
While this capability probably won’t go away entirely, Wccftech, citing unnamed sources, suggests that we could see limitations similar to those imposed on AMD’s Ryzen 7 5800X3D. That chip broke with tradition and only supported undervolting as a means to boost performance.
On Ryzen chips, undervolting can shift the curve used by AMD’s boost algorithm enabling higher clocks at lower temperatures. However, plans for high-end overclocking motherboards in the form of the X670 Extreme chipset cast some doubt on these claims.
The Register reached out to AMD seeking comment on the leaked specs; we’ll let you know if we hear anything. ®
AMD’s Ryzen 7000 desktop processors will reportedly top 5.7 GHz in the case of the Zen giant’s top-of-the-line 7950X, when they launch later this quarter.
The industry watchers at Wccftech claim to have obtained detailed specs for AMD’s next-gen Zen 4 desktop CPUs, codenamed Raphael. If true, they offer some interesting insights into where AMD is going with its silicon designs.
It appears AMD is focusing more on IPC and clock gains in this generation and isn’t as interested in competing with Intel on core count. Here’s a breakdown of the SKUs per the leak:
- Ryzen 9 7950X: 16 core / 32 threads, with base clock of 4.5GHz, a boost clock of 5.7GHz, a TDP of 170W, and 64MB of L3 / 16MB of L2 cache.
- Ryzen 9 7900X: 12 core / 24 threads, with a base clock of 4.7GHz, a boost clock of 5.6GHz, a TDP of 170W, and 64MB of L3 / 12MB of L2 cache.
- Ryzen 7 7700X: 8 core / 16 threads, with a base of clock of 4.5GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 8MB of L2 cache.
- Ryzen 5 7600X: 6 core / 12 threads, with a base clock of 4.7GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 6MB of L2 cache.
All four SKUs will be fabbed using TSMC’s 5nm process. AMD confirmed earlier it will launch 5nm Zen 4 Ryzen 7000 parts this quarter.
TDP lifts clocks
If the leak is correct, the four chips see a notable increase in stated TDP over the 5000-series counterparts they replace. On the high-end, AMD 7000-series Ryzen 9 processors will reportedly suck back an additional 65W compared to the 5950X and 5900X. Meanwhile, AMD’s Ryzen 7 and Ryzen 5 processors have seen their rated TDPs boosted by roughly 40W.
It should be noted that a chip’s actual power consumption will likely be higher as AMD’s stated TDPs don’t actually reflect real-world power consumption. For example, under stock settings, AMD’s Ryzen 5900X routinely pulls between 120-140W of power under load, as long as the chip has adequate thermal headroom. The 5900X can suck down as much as 200W when the chipmaker’s Precision Boost Overdrive — an automated overlocking profile found in the bios — is enabled.
Unless AMD has changed the way it reports TDP since the launch of its 5000-series parts, there’s a good chance the chip designer’s Ryzen 7900 and 7950 parts could near 300W in real-world power consumption, putting them in the same territory as Intel’s 12th-gen chips.
- This credit card-sized PC board can use an Intel Core i7
- AMD boasts of record sales, says 5nm Zen 4 Ryzen 7000 coming this quarter
- Linus Torvalds releases Linux 5.19 – using Asahi on an Arm-powered Mac
- Intel experiences another kind of meltdown
However, the higher TDP appears to have netted AMD a substantial clock improvement over its previous-gen offerings. In fact, in a like-for-like comparison, the entire lineup boasts base clocks at 1 GHz or more over the chips they replace. In many cases, the processors’ base clocks now exceed their 5000-series equivalent’s boost clocks.
Cache clues
The leak also offers some insight into the chiplet architecture underpinning AMD’s Ryzen 7000 processors.
Starting off with the Ryzen 7600X, the package features 32MB of L3 cache — the same as the 5600X — but sees its L2 cache doubled to 6MB, or a full 1MB per core. By extension the 7700X features 8MB of L2.
Based on this, we can conclude that AMD is sticking with the dual chiplet architecture for its Ryzen 9 processors with either eight or six cores per core-complex die as we saw with previous generation Ryzen 5000 chips.
This isn’t terribly surprising as AMD hasn’t increased the core count with this generation of chips, at least not yet.
Will AMD nerf overclocking?
Finally, it’s unclear whether AMD will carry forward its longstanding tradition of unlocking every chip for the purposes of overclocking.
While this capability probably won’t go away entirely, Wccftech, citing unnamed sources, suggests that we could see limitations similar to those imposed on AMD’s Ryzen 7 5800X3D. That chip broke with tradition and only supported undervolting as a means to boost performance.
On Ryzen chips, undervolting can shift the curve used by AMD’s boost algorithm enabling higher clocks at lower temperatures. However, plans for high-end overclocking motherboards in the form of the X670 Extreme chipset cast some doubt on these claims.
The Register reached out to AMD seeking comment on the leaked specs; we’ll let you know if we hear anything. ®
AMD’s Ryzen 7000 desktop processors will reportedly top 5.7 GHz in the case of the Zen giant’s top-of-the-line 7950X, when they launch later this quarter.
The industry watchers at Wccftech claim to have obtained detailed specs for AMD’s next-gen Zen 4 desktop CPUs, codenamed Raphael. If true, they offer some interesting insights into where AMD is going with its silicon designs.
It appears AMD is focusing more on IPC and clock gains in this generation and isn’t as interested in competing with Intel on core count. Here’s a breakdown of the SKUs per the leak:
- Ryzen 9 7950X: 16 core / 32 threads, with base clock of 4.5GHz, a boost clock of 5.7GHz, a TDP of 170W, and 64MB of L3 / 16MB of L2 cache.
- Ryzen 9 7900X: 12 core / 24 threads, with a base clock of 4.7GHz, a boost clock of 5.6GHz, a TDP of 170W, and 64MB of L3 / 12MB of L2 cache.
- Ryzen 7 7700X: 8 core / 16 threads, with a base of clock of 4.5GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 8MB of L2 cache.
- Ryzen 5 7600X: 6 core / 12 threads, with a base clock of 4.7GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 6MB of L2 cache.
All four SKUs will be fabbed using TSMC’s 5nm process. AMD confirmed earlier it will launch 5nm Zen 4 Ryzen 7000 parts this quarter.
TDP lifts clocks
If the leak is correct, the four chips see a notable increase in stated TDP over the 5000-series counterparts they replace. On the high-end, AMD 7000-series Ryzen 9 processors will reportedly suck back an additional 65W compared to the 5950X and 5900X. Meanwhile, AMD’s Ryzen 7 and Ryzen 5 processors have seen their rated TDPs boosted by roughly 40W.
It should be noted that a chip’s actual power consumption will likely be higher as AMD’s stated TDPs don’t actually reflect real-world power consumption. For example, under stock settings, AMD’s Ryzen 5900X routinely pulls between 120-140W of power under load, as long as the chip has adequate thermal headroom. The 5900X can suck down as much as 200W when the chipmaker’s Precision Boost Overdrive — an automated overlocking profile found in the bios — is enabled.
Unless AMD has changed the way it reports TDP since the launch of its 5000-series parts, there’s a good chance the chip designer’s Ryzen 7900 and 7950 parts could near 300W in real-world power consumption, putting them in the same territory as Intel’s 12th-gen chips.
- This credit card-sized PC board can use an Intel Core i7
- AMD boasts of record sales, says 5nm Zen 4 Ryzen 7000 coming this quarter
- Linus Torvalds releases Linux 5.19 – using Asahi on an Arm-powered Mac
- Intel experiences another kind of meltdown
However, the higher TDP appears to have netted AMD a substantial clock improvement over its previous-gen offerings. In fact, in a like-for-like comparison, the entire lineup boasts base clocks at 1 GHz or more over the chips they replace. In many cases, the processors’ base clocks now exceed their 5000-series equivalent’s boost clocks.
Cache clues
The leak also offers some insight into the chiplet architecture underpinning AMD’s Ryzen 7000 processors.
Starting off with the Ryzen 7600X, the package features 32MB of L3 cache — the same as the 5600X — but sees its L2 cache doubled to 6MB, or a full 1MB per core. By extension the 7700X features 8MB of L2.
Based on this, we can conclude that AMD is sticking with the dual chiplet architecture for its Ryzen 9 processors with either eight or six cores per core-complex die as we saw with previous generation Ryzen 5000 chips.
This isn’t terribly surprising as AMD hasn’t increased the core count with this generation of chips, at least not yet.
Will AMD nerf overclocking?
Finally, it’s unclear whether AMD will carry forward its longstanding tradition of unlocking every chip for the purposes of overclocking.
While this capability probably won’t go away entirely, Wccftech, citing unnamed sources, suggests that we could see limitations similar to those imposed on AMD’s Ryzen 7 5800X3D. That chip broke with tradition and only supported undervolting as a means to boost performance.
On Ryzen chips, undervolting can shift the curve used by AMD’s boost algorithm enabling higher clocks at lower temperatures. However, plans for high-end overclocking motherboards in the form of the X670 Extreme chipset cast some doubt on these claims.
The Register reached out to AMD seeking comment on the leaked specs; we’ll let you know if we hear anything. ®
AMD’s Ryzen 7000 desktop processors will reportedly top 5.7 GHz in the case of the Zen giant’s top-of-the-line 7950X, when they launch later this quarter.
The industry watchers at Wccftech claim to have obtained detailed specs for AMD’s next-gen Zen 4 desktop CPUs, codenamed Raphael. If true, they offer some interesting insights into where AMD is going with its silicon designs.
It appears AMD is focusing more on IPC and clock gains in this generation and isn’t as interested in competing with Intel on core count. Here’s a breakdown of the SKUs per the leak:
- Ryzen 9 7950X: 16 core / 32 threads, with base clock of 4.5GHz, a boost clock of 5.7GHz, a TDP of 170W, and 64MB of L3 / 16MB of L2 cache.
- Ryzen 9 7900X: 12 core / 24 threads, with a base clock of 4.7GHz, a boost clock of 5.6GHz, a TDP of 170W, and 64MB of L3 / 12MB of L2 cache.
- Ryzen 7 7700X: 8 core / 16 threads, with a base of clock of 4.5GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 8MB of L2 cache.
- Ryzen 5 7600X: 6 core / 12 threads, with a base clock of 4.7GHz, a boost clock of 5.4GHz, a TDP of 105W, and 32MB of L3 / 6MB of L2 cache.
All four SKUs will be fabbed using TSMC’s 5nm process. AMD confirmed earlier it will launch 5nm Zen 4 Ryzen 7000 parts this quarter.
TDP lifts clocks
If the leak is correct, the four chips see a notable increase in stated TDP over the 5000-series counterparts they replace. On the high-end, AMD 7000-series Ryzen 9 processors will reportedly suck back an additional 65W compared to the 5950X and 5900X. Meanwhile, AMD’s Ryzen 7 and Ryzen 5 processors have seen their rated TDPs boosted by roughly 40W.
It should be noted that a chip’s actual power consumption will likely be higher as AMD’s stated TDPs don’t actually reflect real-world power consumption. For example, under stock settings, AMD’s Ryzen 5900X routinely pulls between 120-140W of power under load, as long as the chip has adequate thermal headroom. The 5900X can suck down as much as 200W when the chipmaker’s Precision Boost Overdrive — an automated overlocking profile found in the bios — is enabled.
Unless AMD has changed the way it reports TDP since the launch of its 5000-series parts, there’s a good chance the chip designer’s Ryzen 7900 and 7950 parts could near 300W in real-world power consumption, putting them in the same territory as Intel’s 12th-gen chips.
- This credit card-sized PC board can use an Intel Core i7
- AMD boasts of record sales, says 5nm Zen 4 Ryzen 7000 coming this quarter
- Linus Torvalds releases Linux 5.19 – using Asahi on an Arm-powered Mac
- Intel experiences another kind of meltdown
However, the higher TDP appears to have netted AMD a substantial clock improvement over its previous-gen offerings. In fact, in a like-for-like comparison, the entire lineup boasts base clocks at 1 GHz or more over the chips they replace. In many cases, the processors’ base clocks now exceed their 5000-series equivalent’s boost clocks.
Cache clues
The leak also offers some insight into the chiplet architecture underpinning AMD’s Ryzen 7000 processors.
Starting off with the Ryzen 7600X, the package features 32MB of L3 cache — the same as the 5600X — but sees its L2 cache doubled to 6MB, or a full 1MB per core. By extension the 7700X features 8MB of L2.
Based on this, we can conclude that AMD is sticking with the dual chiplet architecture for its Ryzen 9 processors with either eight or six cores per core-complex die as we saw with previous generation Ryzen 5000 chips.
This isn’t terribly surprising as AMD hasn’t increased the core count with this generation of chips, at least not yet.
Will AMD nerf overclocking?
Finally, it’s unclear whether AMD will carry forward its longstanding tradition of unlocking every chip for the purposes of overclocking.
While this capability probably won’t go away entirely, Wccftech, citing unnamed sources, suggests that we could see limitations similar to those imposed on AMD’s Ryzen 7 5800X3D. That chip broke with tradition and only supported undervolting as a means to boost performance.
On Ryzen chips, undervolting can shift the curve used by AMD’s boost algorithm enabling higher clocks at lower temperatures. However, plans for high-end overclocking motherboards in the form of the X670 Extreme chipset cast some doubt on these claims.
The Register reached out to AMD seeking comment on the leaked specs; we’ll let you know if we hear anything. ®
