MyTechDailyNews Hardware Check: Nova Lake-S 28-Core PassMark Data Shows Single-Thread Promise and Multi-Thread Questions

Recently, a set of PassMark records attributed to an early Intel desktop sample has entered public view. The listings point to Nova Lake-S and a 28-core configuration. The single-thread results, 5210 points and 5191 points, are the headline numbers. The more useful editorial takeaway, however, is narrower: these entries show promise in light-threaded measurement, while the available multi-thread and system-level marks do not yet support a claim that the sample defeats current flagship desktop processors. For hardware buyers, the value of this leak is not a verdict; it is a set of clues about where Intel may still be tuning the platform.

How the sample appeared in PassMark

A VideoCardz reader discovered the entries in PerformanceTest 11.1, specifically inside the Manage Baselines interface. The processor was displayed as Intel 0000, while the database also exposes the short identifier 0000. The earliest traceable timestamp in the report reaches September this year. Two records for a 28-core part were captured. This is why MyTechDailyNews Hardware treats the numbers as preliminary: an unnamed sample can be tested under non-retail power targets, cooling, memory configuration, or firmware settings.

Single-thread table: where the sample looks strong

Comparison item	Score	Relationship to the sample
Nova Lake-S sample, single-thread	5210 points	Lead value used in the cited comparison
Core Ultra 9 285K, single-thread	5059 points	about 3% behind
Ryzen 9 9950X3D, single-thread	4805 points	8.4% behind
Core i9-14900KF, single-thread	4635 points	12.4% behind

The table mirrors the media-selected comparison scores. The sample leads the Core Ultra 9 285K entry by about 3%, the Ryzen 9 9950X3D entry by 8.4%, and the Core i9-14900KF entry by 12.4%. The second 28-core result, 5191 points, sits only slightly below the best single-thread mark shown. Without published test conditions, that small distance is better read as a data band than as a definitive ranking.

Multi-thread table: the part that does not align with the headline

Processor / sample	CPU Mark result	Buyer-facing interpretation
Nova Lake-S sample	61509 points	Best available multi-thread score in this listing
Core Ultra 9 285K	67274 points	Above the sample in the cited data
Ryzen 9 9950X3D	72066 points	Further above the sample in the cited data
Core i9-14900KF	58074 points	Below the sample in the cited data

The CPU Mark picture is more cautious. The sample's highest shown result is 61509 points. That leaves it below Core Ultra 9 285K's 67274 points and below Ryzen 9 9950X3D's 72066 points. Its clear win is only against Core i9-14900KF's 58074 points. For a 28-core desktop processor, that is the most important hardware signal: a single-thread lead does not automatically become an all-core throughput lead.

Memory Mark is a system signal, not a CPU verdict

The PassMark Memory Mark value for the sample is 2523.4. The comparison system listed 3349.6. The difference should not be read as a pure processor failure. Memory Mark is a whole-system item; it can be influenced by RAM capacity, memory timing, platform settings, BIOS, cooling, and the specific test baseline. The sample screenshot also showed about 16GB memory, about 3.3GHz reported frequency, and near 4.8GHz turbo read. Those values help describe the test machine, but they do not identify the retail configuration or final memory support.

The 8-core entries add another caution

The same database also contains 8-core records. The best single-thread mark in that group is 5085 points, about 0.5% above the Core Ultra 9 285K comparison value. The CPU Mark results, however, range from 9004 to 23478 points. That spread is much wider than a normal retail comparison table would suggest. It points toward inconsistent test conditions, sample states, or development settings. If a reader wants to use these records as a planning reference, the 8-core group should be treated as noisier than the 28-core group.

Product-table leaks: naming, cache, and cores

• A previously leaked Nova Lake-S product table lists models such as Core Ultra 9 4970K BFC.
• The high-end configuration is described as using 8 performance cores, 16 efficient cores, and 4 low-power efficient cores.
• The BFC suffix may represent a larger last-level cache version, but the benchmark entry does not confirm whether the tested sample carries that extra cache.
• The standard version 4950K may also include the 4 low-power efficient cores, although the leaked information does not identify the exact retail SKU behind the PassMark result.

Platform and timing: what is still unconfirmed

1. Nova Lake-S is rumored to move to an LGA 1954 socket.
2. The platform is tied to a 900-series chipset in current leak narratives.
3. Intel has confirmed that Nova Lake is planned for before end of 2026.
4. Media expectations place first desktop products at early 2027.
5. A version with up to 52-core dual compute chip is discussed for a later point, but remains subject to official confirmation.

For MyTechDailyNews Hardware, the practical meaning is platform risk. Socket, chipset, SKU naming, cache variant, and launch timing are still not fully established. A procurement or upgrade plan built only around these database entries would be premature.

Common mistakes in reading this leak

• Treating a single-thread result as proof that the processor will outperform every flagship in games, productivity, or workstation work.
• Comparing CPU Mark results without knowing whether the sample used retail power limits, retail cooling, or comparable BIOS settings.
• Reading Memory Mark as a CPU-only benchmark when it reflects the tested machine.
• Assuming that BFC, 4970K, or 4950K labels from leak tables map cleanly to final retail models.
• Using the 8-core score range as if all samples were validated under the same conditions.

Action advice by reader type

Reader type	What the data supports	What the data does not support	Recommended action
Hardware enthusiast	Watch whether Nova Lake-S keeps a single-thread advantage once retail clocks and power targets are known.	Do not treat the sample as a confirmed flagship replacement.	Track follow-up PassMark or validated review data before drawing conclusions.
Gaming PC buyer	High single-thread scores are interesting, but only if stable on retail platform.	The leak does not prove higher in-game frame rates or lower latency.	Wait for independent game and platform testing.
Content creator	Multi-thread is usually more relevant for rendering and compilation.	The sample trails two cited comparators in CPU Mark.	Do not plan upgrades around this result alone; wait for retail multi-thread benchmarks.
IT procurement	New platform indicators include LGA 1954 socket and 900-series chipset.	Official SKU availability, validation, and timing are not fully confirmed.	Hold vendor roadmap decisions until Intel confirms launch details and tested configurations.

Conclusion: the leak is a checkpoint, not a certificate

These PassMark records are useful because they put Nova Lake-S into a measurable timeline. The two 28-core entries show 5210 points and 5191 points in single-thread work, while the best CPU Mark result of 61509 points still sits below the listed 67274 and 72066 point comparators. Memory Mark at 2523.4 versus 3349.6 adds another reason to be careful: the machine, not only the chip, shapes that number. The product-table hints around Core Ultra 9 4970K BFC, 4950K, 8 performance cores, 16 efficient cores, and 4 low-power efficient cores give the leak more structure, but they still do not identify the exact retail model or cache version behind the result. For MyTechDailyNews Hardware, the best next step is to watch for official confirmation of the LGA 1954 socket, 900-series chipset, and the final schedule after before end of 2026 and the widely expected early 2027 desktop window, including any later 52-core dual compute chip option. Until then, the data is a clue for enthusiasts and planners, not a purchase certificate.