🇷🇺 Русский · 🇬🇧 English · 🇩🇪 Deutsch · 🇫🇷 Français · 🇪🇸 Español · 🇮🇹 Italiano

Comparison with competitors

All tables below are generated automatically from honest comparative tests (benchmarks/compare.php) running the SAME operation for every competitor, PHP 8.3.32, without Xdebug.

1. Features

Capability	🏆 CloudCastle	symfony	carbon	lcobucci	beste	ergebnis	new¹
PSR-20 ClockInterface (now(): DateTimeImmutable)	✅	✅	✅	✅	✅	✅	❌
Frozen clock for deterministic tests	✅	✅	✅	✅	✅	✅	❌
Mock clock with virtual sleep()	✅	✅	❌	❌	❌	❌	❌
Monotonic clock (immune to system time jumps)	✅	✅	❌	❌	❌	❌	❌
Time offset decorator (OffsetClock)	✅	❌	❌	❌	❌	❌	❌
Time scale decorator (ScaledClock)	✅	❌	❌	❌	❌	❌	❌
Truncation to an arbitrary unit	✅	❌	❌	❌	❌	❌	❌
Duration value object with nanosecond precision	✅	❌	✅	❌	❌	❌	❌
Deadlines with remaining time (Deadline)	✅	❌	❌	❌	❌	❌	❌
Step and queue clocks for scenario tests	✅	❌	❌	❌	❌	❌	❌
Stopwatch with laps included in the package	✅	❌	❌	❌	❌	❌	❌
Cheap timestamp()/microtime() primitives without date objects	✅	❌	❌	❌	❌	❌	❌
Application-clock-bound moment (DatePoint)	✅	✅	✅	❌	❌	❌	❌
Scoped global clock substitution with auto-restore	✅	❌	✅	❌	❌	❌	❌
Adapter for any PSR-20 clock	✅	✅	❌	❌	✅	❌	❌
Calendar month/year arithmetic with overflow control (CalendarDate)	✅	❌	✅	❌	❌	❌	❌
Localized diffForHumans and isoFormat without ext-intl	✅	❌	❌	❌	❌	❌	❌
Calendar predicates and diffs (isWeekend, diffInDays, next)	✅	❌	✅	❌	❌	❌	❌
PHPUnit trait for mocking time (mockTime/unmockTime)	✅	✅	❌	❌	❌	❌	❌
Real sleep() on system and monotonic clocks	✅	✅	❌	❌	❌	❌	❌
Wrapping any object with a now() method as PSR-20	✅	❌	❌	❌	✅	❌	❌
Calendar period iteration with a step (CalendarPeriod)	✅	❌	✅	❌	❌	❌	❌
Human-readable duration (TimeSpan::forHumans)	✅	❌	✅	❌	❌	❌	❌
Macros: extending the date API without inheritance	✅	❌	✅	❌	❌	❌	❌
Total	🏆 24	8	10	2	4	2	0

2. Security and correctness

Property	🏆 CloudCastle	symfony	carbon	lcobucci	beste	ergebnis	new¹
Immutable readings (DateTimeImmutable)	✅	✅	✅	✅	✅	✅	✅
Explicit timezone (independent of date.timezone)	✅	✅	✅	✅	❌	✅	❌
Test determinism without mandatory global state	✅	✅	❌	✅	✅	✅	❌
Monotonicity (resilient to system clock rollback)	✅	✅	❌	❌	❌	❌	❌
Typed exceptions instead of false/warnings at the input boundary	✅	❌	✅	❌	❌	❌	❌
Hostile input sanitising in exception messages	✅	❌	❌	❌	❌	❌	❌
Integer overflow control for durations	✅	❌	❌	❌	❌	❌	❌
Minimal dependencies (≤2, without ICU/ext-intl)	✅	❌	❌	✅	✅	✅	✅
Total	🏆 8	4	3	4	3	4	2

2.1. Compliance with Composer package security standards

Standard	CloudCastle	symfony	carbon	lcobucci	beste	ergebnis	🏆 Winners
Security policy (SECURITY.md or a published process)	✅	✅	✅	❌	✅	✅	CloudCastle, symfony, carbon, beste, ergebnis
Known-vulnerability control for dependencies (roave/security-advisories + composer audit)	✅	❌	❌	❌	❌	❌	CloudCastle
OpenSSF Scorecard: published score ≥ 7/10	❌	✅	❌	❌	❌	❌	symfony
CWE-20: input validation with typed exceptions	✅	❌	✅	❌	❌	❌	CloudCastle, carbon
CWE-117/CWE-150: neutralising control characters in error messages	✅	❌	❌	❌	❌	❌	CloudCastle
CWE-190: integer overflow control	✅	❌	❌	❌	❌	❌	CloudCastle
Supply chain: ≤2 runtime dependencies without polyfills/ICU	✅	❌	❌	✅	✅	✅	CloudCastle, lcobucci, beste, ergebnis
OpenSSF Best Practices: automated tests with mutation control in CI	✅	❌	❌	✅	❌	✅	CloudCastle, lcobucci, ergebnis
OSI-approved licence in the distribution (MIT)	✅	✅	✅	✅	✅	✅	CloudCastle, symfony, carbon, lcobucci, beste, ergebnis
Total	🏆 8	3	3	3	3	4	CloudCastle

<sub>Sources: SECURITY.md files in the repositories (GitHub API), require-dev of the distributions on Packagist, api.securityscorecards.dev (symfony/clock — score of the symfony/symfony monorepo; OpenSSF Scorecard rates GitHub projects only, so CloudCastle on GitVerse has no score — an honest platform minus); CWE rows — from the packages’ security tests; symfony/clock for PHP 8.1 is the 6.4 branch with polyfill-php83.</sub>

3. Performance: now()

_Fetching the current moment now(), 200 000 times (min of 15)._

Solution	Time (ms)	Verdict
new¹	109,7	baseline (not a library)
🏆 CloudCastle	114,5	fastest among libraries
🏆 ergebnis	115,6	on par with the leader (within measurement error)
🏆 lcobucci	116	on par with the leader (within measurement error)
🏆 beste	116,8	on par with the leader (within measurement error)
symfony	154,6	competitor
carbon	649,8	competitor

4. Performance: unix timestamp

_Current unix timestamp (whole seconds), 200 000 times._

Solution	Time (ms)	Verdict
new¹	16	baseline (not a library)
🏆 CloudCastle	27,8	fastest among libraries
ergebnis	122,3	competitor
lcobucci	123	competitor
beste	125,6	competitor
symfony	164,2	competitor
carbon	658,7	competitor

5. Library memory consumption

_Library-only memory growth over 200 000 now() calls (isolated process, without autoloader weight)._

Solution	Memory (KB)	Verdict
new¹	1	baseline (not a library)
🏆 beste	3	lightest among libraries
lcobucci	5	competitor
ergebnis	5	competitor
symfony	30	competitor
CloudCastle	31	competitor
carbon	3 425	competitor

6. Memory leaks

_Memory growth after warm-up over 200 000 now() calls (0 — no leaks)._

Solution	Memory (KB)	Verdict
🏆 CloudCastle	0	leak-free
symfony	0	leak-free
carbon	0	leak-free
lcobucci	0	leak-free
beste	0	leak-free
ergebnis	0	leak-free
new¹	0	baseline (not a library)

7. Process memory peak

_Peak memory of an isolated process over 200 000 now() calls._

Solution	Memory (KB)	Verdict
🏆 CloudCastle	7 035	lightest among libraries
🏆 symfony	7 035	on par with the leader (within measurement error)
🏆 lcobucci	7 035	on par with the leader (within measurement error)
🏆 beste	7 035	on par with the leader (within measurement error)
🏆 ergebnis	7 035	on par with the leader (within measurement error)
new¹	7 035	baseline (not a library)
carbon	10 303	competitor

8. Code quality

Tool	🏆 CloudCastle	symfony	carbon	lcobucci	beste	ergebnis
PHPStan	✅ max + strict	❌	✅	✅ + strict	✅ + strict	✅ + strict
Psalm	✅ level 1	✅ (monorepo)	❌	❌	✅	❌
PHPMD	✅	❌	✅	❌	❌	❌
PHPCS / CS-Fixer	✅ PSR-12 + fixer	✅ fixer	✅	✅	❌	✅ fixer
Rector	✅	❌	❌	❌	❌	✅
Deptrac (architecture layers)	✅	❌	❌	❌	❌	❌
Mutation testing (Infection)	✅ MSI 100.0%	❌	❌	✅	❌	✅
Line coverage (live)	✅ 100.0%	—	—	—	—	—
Total	🏆 8	2	3	3	2	4

<sub>Competitor tooling is taken from the require-dev of their published distributions (snapshot July 2026); symfony/clock — from the symfony/symfony monorepo. CloudCastle metrics are live local-run results.</sub>

¹ The baseline (native calls without the completeness of a solution) is shown for context and does not compete with the libraries.

Results within 5% of the best are marked as equal: these solutions share a character-identical hot path (new DateTimeImmutable('now', $tz)), so at that scale the difference comes from machine noise, not code.

🇷🇺 Русский · 🇬🇧 English · 🇩🇪 Deutsch · 🇫🇷 Français · 🇪🇸 Español · 🇮🇹 Italiano